INTEGRATED CIRCUITS DATA SHEET SAA4956TJ 2.9-Mbit field memory with noise reduction Preliminary specification File under Integrated Circuits, IC02 1998 Dec 08 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction CONTENTS 1 FEATURES 2 GENERAL DESCRIPTION 3 ORDERING INFORMATION 4 QUICK REFERENCE DATA 5 BLOCK DIAGRAM 6 PINNING 7 FUNCTIONAL DESCRIPTION 7.1 7.1.1 7.1.2 7.1.3 7.1.4 7.1.5 7.1.6 7.1.7 7.2 7.2.1 7.2.2 7.2.3 7.2.4 7.2.5 7.3 Field memory function Write operation Read operation Power-up and initialization Old and new data access Memory arbitration logic and self-refresh Cascade operation Test mode operation Noise reduction function Reformatting and formatting Band-splitting Motion detection K-factor Noise shape I2C-bus interface 1998 Dec 08 8 LIMITING VALUES 9 THERMAL CHARACTERISTICS 10 CHARACTERISTICS 11 APPLICATION INFORMATION 12 PACKAGE OUTLINE 13 SOLDERING 13.1 Introduction to soldering surface mount packages Reflow soldering Wave soldering Manual soldering Suitability of surface mount IC packages for wave and reflow soldering methods 13.2 13.3 13.4 13.5 2 SAA4956TJ 14 DEFINITIONS 15 LIFE SUPPORT APPLICATIONS 16 PURCHASE OF PHILIPS I2C COMPONENTS Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 1 SAA4956TJ FEATURES • 2949264-bit field memory with optional field based noise reduction • 245772 × 12-bit organization • 3.3 V power supply • Inputs fully TTL compatible when using an extra 5 V power supply However, the SAA4956TJ has also, in addition to the field memory function, a field based noise reduction circuit. If this function is enabled it can be controlled via the I2C-bus. • High speed read and write operations • FIFO operations: The maximum storage depth is 245772 words × 12 bits. A FIFO operation with full word continuous read and write could be used as a data delay, for example. A FIFO operation with asynchronous read and write could be used as a data rate multiplier. Here the data is written once, then read as many times as required as long as new data is not written. In addition to the FIFO operations, a random block access mode is accessible during the pointer reset operation. When this mode is enabled, reading and/or writing may begin at, or proceed from, the start address of any of the 6144 blocks. Each block is 40 words in length. Two or more SAA4956TJs can be cascaded to provide a greater storage depth or a longer delay, without the need for additional circuitry. – Full word continuous read and write – Independent read and write pointers (asynchronous read and write access) – Resettable read and write pointers. • Optional field based noise reduction activated by an enable pin and controlled via the I2C-bus • Optional random access by block function (40 words per block) enabled during pointer reset operation • Quasi static (internal self-refresh and clocking pauses of infinite length) • Write mask function • Cascade operation possible The SAA4956TJ contains separate 12-bit wide serial ports for reading and writing. The ports are controlled and clocked separately, so asynchronous read and write operations are supported. Independent read and write clock rates are possible. Addressing is controlled by read and write address pointers. Before a controlled write operation can begin, the write pointer must be set to zero or to the beginning of a valid address block. Likewise, the read pointer must be set to zero or to the beginning of a valid address block before a controlled read operation can begin. • Compatible with SAA4955TJ • 16-Mbit CMOS DRAM process technology • 40-pin SOJ package. 2 GENERAL DESCRIPTION The SAA4956TJ is a 2949264-bit field memory with an optional field based noise reduction designed for advanced TV applications such as 100/120 Hz TV, PALplus, PIP and 3D comb filter. The SAA4956TJ is functional and pin compatible with the SAA4955TJ. 3 ORDERING INFORMATION PACKAGE TYPE NUMBER NAME SAA4956TJ SOJ40 1998 Dec 08 DESCRIPTION plastic small outline package; 40 leads (J-bent); body width 10.16 mm 3 VERSION SOT449-1 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 4 SAA4956TJ QUICK REFERENCE DATA SYMBOL Tcy(SWCK) PARAMETER SWCK cycle time CONDITIONS MIN. TYP. MAX. UNIT NREN = LOW; see Fig.4 26 − − ns NREN = HIGH; see Fig.4 52 − 150 ns Tcy(SRCK) read cycle time (SRCK) see Fig.11 26 − − ns see Fig.11 tACC read access time after SRCK − − 21 ns VDD supply voltage (pin 19) 3.0 3.3 3.6 V VDD(O) supply voltage (pin 22) 3.0 3.3 3.6 V VDD(P) supply voltage (pin 21) 3.0 3.3 5.5 V IDD(tot) total supply current (IDD(tot) = IDD + IDD(O) + IDD(P)) − 27 70 mA 1998 Dec 08 minimum read/write cycle; outputs open 4 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 5 SAA4956TJ BLOCK DIAGRAM D0(V0) to D11(Y7) handbook, full pagewidth IE 12 14 to 3 NREN SCL 40 1 18 DATA INPUT AND WRITE MASK BUFFER (×13) WE SDA RSTW SWCK 17 20 16 15 I2C-BUS INTERFACE SAA4956TJ IE internal 12 + 1 D0 internal IE internal I2C-bus control NOISE REDUCTION DATA MUX INPUT BUFFER (×3) 3 SERIAL WRITE/READ2 CONTROLLER mini cache write/read2 control + cache transfer 12 + 1 12 SERIAL WRITE REGISTER 20-WORD (×13) D-field delay load write block address D0 internal PARALLEL WRITE REGISTER 20-WORD (×13) 12 write acknowledge read2 acknowledge read2 control 20 × (12 + 1) DATA MUX 12 write control IE internal 20 × (12 + 1) 12 + 1 WRITE ADDRESS COUNTER WRITE REGISTER MINI CACHE 12-WORD (×12) READ2 ADDRESS COUNTER MEMORY ARBITRATION REFRESH ADDRESS LOGIC COUNTER MEMORY ARRAY 245760-WORD (×12) READ2 READ REGISTER REGISTER READ ADDRESS COUNTER 20 × 12 20 × 12 PARALLEL READ2 REGISTER 20-WORD (×12) PARALLEL READ REGISTER 20-WORD (×12) 20 × 12 +3.3 V VDD 19 VDD(P) 21 VDD(O) 22 GND CLOCK OSCILLATOR SERIAL READ REGISTER 20-WORD (×12) read control 12 read acknowledge load read block address mini cache read control 100 nF OGND refresh clock 20 × 12 SERIAL READ2 REGISTER 20-WORD (×12) 12 OE internal DATA MUX 2 39 DATA OUTPUT BUFFER (×12) 27 to 38 OE internal 3 INPUT BUFFER (×4) 23 24 25 SERIAL READ CONTROLLER 26 MGR687 12 Q0(V0) to Q11(Y7) OE RE RSTR SRCK Pin 21 (VDD(P)) should be connected to the supply voltage of the driving circuit that generates the input voltages. This could be, for instance, 5.0 V instead of 3.3 V. Pins 19 and 22 (VDD and VDD(O)) require a 3.3 V supply. Fig.1 Block diagram. 1998 Dec 08 5 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 6 SAA4956TJ PINNING SYMBOL PIN I/O DESCRIPTION I2C-bus SCL 1 digital input serial clock of GND 2 ground general purpose ground D11(Y7) 3 digital input data input 11, Y input bit 7 if NREN is HIGH D10(Y6) 4 digital input data input 10, Y input bit 6 if NREN is HIGH D9(Y5) 5 digital input data input 9, Y input bit 5 if NREN is HIGH D8(Y4) 6 digital input data input 8, Y input bit 4 if NREN is HIGH D7(Y3) 7 digital input data input 7, Y input bit 3 if NREN is HIGH D6(Y2) 8 digital input data input 6, Y input bit 2 if NREN is HIGH D5(Y1) 9 digital input data input 5, Y input bit 1 if NREN is HIGH D4(Y0) 10 digital input data input 4, Y input bit 0 if NREN is HIGH D3(U1) 11 digital input data input 3, U input bits 1, 3, 5, 7 if NREN is HIGH D2(U0) 12 digital input data input 2, U input bits 0, 2, 4, 6 if NREN is HIGH D1(V1) 13 digital input data input 1, V input bits 1, 3, 5, 7 if NREN is HIGH D0(V0) 14 digital input data input 0, V input bits 0, 2, 4, 6 if NREN is HIGH SWCK 15 digital input serial write clock RSTW 16 digital input write reset clock WE 17 digital input write enable IE 18 digital input input enable VDD 19 supply 3.3 V general purpose supply voltage SDA 20 digital I/O serial data of I2C-bus VDD(P) 21 supply 3.3 to 5.5 V supply voltage for protection circuits VDD(O) 22 supply 3.3 V supply voltage for output circuits OE 23 digital input output enable RE 24 digital input read enable RSTR 25 digital input reset read SRCK 26 digital input serial read clock Q0(V0) 27 digital output data output 0, V output bits 0, 2, 4, 6 if NREN is HIGH Q1(V1) 28 digital output data output 1, V output bits 1, 3, 5, 7 if NREN is HIGH Q2(U0) 29 digital output data output 2, U output bits 0, 2, 4, 6 if NREN is HIGH Q3(U1) 30 digital output data output 3, U output bits 1, 3, 5, 7 if NREN is HIGH Q4(Y0) 31 digital output data output 4, Y output bit 0 if NREN is HIGH Q5(Y1) 32 digital output data output 5, Y output bit 1 if NREN is HIGH Q6(Y2) 33 digital output data output 6, Y output bit 2 if NREN is HIGH Q7(Y3) 34 digital output data output 7, Y output bit 3 if NREN is HIGH Q8(Y4) 35 digital output data output 8, Y output bit 4 if NREN is HIGH Q9(Y5) 36 digital output data output 9, Y output bit 5 if NREN is HIGH Q10(Y6) 37 digital output data output 10, Y output bit 6 if NREN is HIGH Q11(Y7) 38 digital output data output 11, Y output bit 7 if NREN is HIGH OGND 39 ground ground for output circuits NREN 40 digital input noise reduction enable 1998 Dec 08 6 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction the read operation frequency. In this case the random block access modes are not supported because a second read operation (READ2) is activated with an identical frequency as used in the write operations. The PAN-IC (SAA4995WP) needs approximately the same write frequency for the noise reduction option as the read frequency (32 MHz). To allow this configuration the self-refresh must be switched off via the I2C-bus interface. handbook, halfpage SCL 1 40 NREN GND 2 39 OGND D11(Y7) 3 38 Q11(Y7) D10(Y6) 4 37 Q10(Y6) D9(Y5) 5 36 Q9(Y5) D8(Y4) 6 35 Q8(Y4) D7(Y3) 7 34 Q7(Y3) D6(Y2) 8 33 Q6(Y2) D5(Y1) 9 32 Q5(Y1) D4(Y0) 10 31 Q4(Y0) 7.1.1 29 Q2(U0) D1(V1) 13 28 Q1(V1) D0(V0) 14 27 Q0(V0) SWCK 15 26 SRCK RSTW 16 25 RSTR WE 17 24 RE IE 18 23 OE VDD 19 22 VDD(O) SDA 20 21 VDD(P) 7.1.1.1 RSTW needs to stay LOW for a single SWCK cycle before another reset write operation can take place. If RSTW is HIGH for 1024 SWCK write clock cycles while WE is HIGH, the SAA4956TJ will enter a built-in test mode. Fig.2 Pin configuration. 7.1.1.2 FUNCTIONAL DESCRIPTION • The basic field memory function (see Section 7.1) • The optional noise reduction function (used in case the NREN signal is HIGH; see Section 7.2) 1. At the 1st and 2nd positive transitions of SWCK, IE must be LOW and WE must be HIGH • The I2C-bus interface function (which controls the noise reduction circuit; see Section 7.3). 2. At the 3rd and 4th positive transitions of SWCK, IE must be HIGH and WE must be LOW 3. At the 5th positive transition of SWCK, the state of WE determines which input pin is used for the block address. If WE is LOW the Most Significant Bit (MSB) of the block address must be applied to the D0 input pin. If WE is HIGH, the MSB of the block address is applied to pin IE. Field memory function The basic field memory function is fully compatible with the SAA4955TJ if the NREN signal is LOW. In this case the noise reduction function is bypassed via a data mux. If the NREN signal is HIGH the basic field memory function can only be executed with a write frequency restricted to half of 1998 Dec 08 Random write block access mode The SAA4956TJ will enter random write block access mode if the following signal sequence is applied to control inputs IE and WE during the first four SWCK write clock cycles after a reset write (see Figs 6 and 7): The functional description is divided into 3 main sections: 7.1 Reset write: RSTW The first positive transition of SWCK after RSTW goes from LOW-to-HIGH resets the write address pointer to the lowest address (−12 decimal), regardless of the state of WE (see Figs 4 and 5). RSTW set-up (tsu(RSTW)) and hold (th(RSTW)) times are referenced to the rising edge of SWCK (see Fig.4). The reset write operation may also be asynchronously related to the SWCK signal if WE is LOW. MGR688 7 WRITE OPERATION Write operations are controlled by the SWCK, RSTW, WE and IE signals. A write operation starts with a reset write address pointer (RSTW) operation, followed by a complete cycle of the SWCK clock during which time WE and IE must be held HIGH. Write operations between two successive reset write operations must contain at least 40 SWCK write clock cycles while WE is HIGH. To transfer data temporarily stored in the serial write registers to the memory array, a reset write operation is required after the last write operation. D3(U1) 11 SAA4956TJ 30 Q3(U1) D2(U0) 12 SAA4956TJ 7 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction During this time, control signals WE and IE will function as defined for normal operation. The remaining 12 bits of the 13-bit write block address must be applied, in turn, to the selected input pin (D0 or IE) at the following 12 positive transitions of SWCK. The Least Significant Bit (LSB) of the write block address is applied at the 17th positive transition of SWCK. A write latency period of 18 additional SWCK clock cycles is required before write access to the new block address is possible. During this time, data is transferred from the serial write and parallel write registers into the memory array and the write pointer is set to the new block address. The address area reserved for the mini cache, accessible after a standard reset operation, is from decimal −12 to −1. The memory array starts at decimal 0 and ends at 245759. Decimal address 0 is identical to block address 0000H. Because a single block address is defined for every 40 words in the memory array, block address 0001H corresponds to decimal address 40. The highest block address is 17FFH. This block has a decimal start address of 245720 and an end address of 245759. If a read or write reset operation is not performed, the next read or write pointer address after 245759 will be address 0 due to pointer wraparound. It should be noted that reset read and write operations should occur together. If one pointer wraps around while the other is reset, either 12 words will be lost or 12 words of undefined data will be read. Block address values between 0 and 6143 are valid. Values outside this range must be avoided because invalid block addresses can result in abnormal operation or a lock-up condition. Recovery from lock-up requires a standard reset write operation. 7.1.1.4 WE must remain LOW from the 3rd positive transition of SWCK to the 17th write latency SWCK clock cycle if the block address is applied to pin D0. If the block address is applied to pin IE, WE must be HIGH on the 5th positive transition of SWCK, may be HIGH or LOW on the 6th transition, and must be LOW from the 7th transition to the 17th write latency SWCK clock cycle. 7.1.1.5 Write enable: WE Pin WE is used to enable or disable a data write operation. The WE signal controls data inputs D0 to D11. In addition, the internal write address pointer is incremented if WE is HIGH at the positive transition of the SWCK write clock. WE set-up (tsu(WE)) and hold (th(WE)) times are referenced to the positive edge of SWCK (see Fig.8). The LOW-to-HIGH transition on RSTW required at the beginning of the sequence should not be repeated. Additional LOW-to-HIGH transitions on RSTW would disable write block address mode and reset the write pointer. 7.1.1.6 Address organization Input enable: IE Pin IE is used to enable or disable a data write operation from the D0 to D11 data inputs into memory. The latched data will only be written into memory if the IE and WE signals were HIGH during the previous positive transition of SWCK. A LOW level on IE will prevent the data being written into memory and existing data will not be overwritten (write mask function; see Fig.10). The IE set-up (tsu(IE)) and hold (th(IE)) times are referenced to the positive edge of SWCK (see Fig.9). Two different types of memory are used in the data address area: a mini cache for the first 12 data words after a reset write or a reset read, and a DRAM cell memory array with a 245760 word capacity. Each word is 12 bits long. The mini cache is needed to store data temporarily after a reset operation since a latency period is required before read or write access to the memory array is possible. Latency periods are needed for read or write operations in random read or write block access modes because data is read from, or written to, the memory array. The data in the mini cache can only be accessed directly after a standard reset operation. It cannot be accessed in random read or write block access modes. 1998 Dec 08 Data inputs: D0 to D11 and write clock: SWCK A positive transition on the SWCK write clock latches the data on inputs D0 to D11, provided WE was HIGH at the previous positive transition of SWCK. The data input set-up (tsu(D)) and hold (th(D)) times are referenced to the positive transition of SWCK (see Fig.5). The latched data will only be written into memory if IE was HIGH at the previous positive transition of SWCK. At the 18th write latency SWCK clock cycle, IE and WE may be switched HIGH to prepare for writing new data at the next positive transition of SWCK. The complete write block access entry sequence is finished after the 18th write latency cycle. 7.1.1.3 SAA4956TJ 8 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 7.1.2 The data output pins are not controlled by the OE pin and are forced into high impedance mode from the 3rd to the 17th positive transition of SRCK. OE should be held LOW during the read latency period. RE must remain LOW from the 3rd positive transition of SRCK to the 20th read latency SRCK clock cycle. READ OPERATION Read operations are controlled by the SRCK, RSTR, RE and OE signals. A read operation starts with a reset read address pointer (RSTR) operation, followed by a complete cycle of the SRCK clock during which time RE and OE must be held HIGH. Read operations between two successive reset read operations must contain at least 20 SRCK read clock cycles while RE is HIGH. 7.1.2.1 After the 20th read latency SRCK clock cycle, RE and OE may be switched HIGH to prepare for reading new data from the new address block at the next positive transition of SRCK. The complete read block access entry sequence is finished after the 20th read latency cycle. Reset read: RSTR The first positive transition of SRCK after RSTR goes from LOW-to-HIGH resets the read address pointer to the lowest address (−12 decimal; see Figs 11 and 12). If RE is LOW, however, the reset read operation to the lowest address will be delayed until the first positive transition of SRCK after RE goes HIGH. RSTR set-up (tsu(RSTR)) and hold (th(RSTR)) times are referenced to the rising edge of SRCK (see Fig.11). The reset read operation may also be asynchronously related to the SRCK signal if RE is LOW. The LOW-to-HIGH transition on RSTR required at the beginning of the sequence should not be repeated. Additional LOW-to-HIGH transitions on RSTR would disable the read block address mode and reset the read pointer. 7.1.2.3 RSTR needs to stay LOW for a single SRCK cycle before another reset write operation can take place. 7.1.2.2 Data outputs: Q0 to Q11 and read clock: SRCK The new data is shifted out of the data output registers on the rising edge of the SRCK read clock provided RE and OE are HIGH. Data output pins are low impedance if OE is HIGH. If OE is LOW, the data outputs are high impedance and the data output bus may be used by other devices. Data output hold (th(Q)) and access times (tACC) are referenced to the positive transition of SRCK. The output data becomes valid after access time interval tACC (see Fig.12). Random read block access mode The SAA4956TJ will enter random read block access mode if the following signal sequence is applied to control inputs RE and OE during the first four SWCK write clock cycles after a reset read (see Fig.13): 1. At the 1st and 2nd positive transitions of SRCK, OE must be LOW and RE must be HIGH Data output pins Q0 to Q11 are TTL compatible with the restriction that when the outputs are high impedance, they must not be forced higher than VDD(O) + 0.5 V or 5.0 V absolute. The output data has the same polarity as the incoming data at inputs D0 to D11. 2. At the 3rd and 4th positive transitions of SRCK, OE must be HIGH and RE must be LOW. During this time, control signals RE and OE will function as defined for normal operation. The Most Significant Bit (MSB) of the block read address is applied to the OE input pin at the 5th positive transition of SRCK. The remaining 12 bits of the 13-bit read block address must be applied, in turn, to OE at the following 12 positive transitions of SWCK. The Least Significant Bit (LSB) of the block address is applied at the 17th positive transition of SRCK. A read latency period of 20 additional SRCK clock cycles is required before read access to the new block address is possible. During this period, data is transferred from the memory array to the serial read and parallel read registers and the read pointer is set to the new block address. 7.1.2.4 Read enable: RE RE is used to increment the read pointer. Therefore, RE needs to be HIGH at the positive transition of SRCK. When RE is LOW, the read pointer is not incremented. RE set-up (tsu(RE)) and hold times (th(RE)) are referenced to the positive edge of SRCK (see Fig.14). 7.1.2.5 Output enable: OE OE is used to enable or disable data outputs Q0 to Q11. The data outputs are enabled (low impedance) if OE is HIGH. OE LOW disables the data output pins (high impedance). Incrementing of the read pointer does not depend on the status of OE. OE set-up (tsu(OE)) and hold times (th(OE)) are referenced to the positive edge of SRCK (see Fig.15). Block address values between 0 and 6143 are valid. Values outside this range must be avoided because invalid block addresses can result in abnormal operation or a lock-up condition. Recovery from lock-up requires a standard reset read operation. 1998 Dec 08 SAA4956TJ 9 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 7.1.3 POWER-UP AND INITIALIZATION 7.1.5 Reliable operation is not guaranteed until at least 100 µs after power-up, the time needed to stabilize VDD within the recommended operating range. After the 100 µs power-up interval has elapsed, the following initialization sequence must be performed: a minimum of 12 dummy read operations (SRCK cycles) followed by a reset read operation (RSTR), and a minimum of 12 dummy write operations (SWCK) followed by a reset write operation (RSTW). Read and write initialization may be performed simultaneously. 7.1.6 CASCADE OPERATION If a longer delay is needed, the total storage depth can be increased beyond 2949264 bits by cascading several SAA4956TJs. For details see the interconnection and timing diagrams (Figs 18 and 19). OLD AND NEW DATA ACCESS The noise reduction function can be realized by enabling this function with the NREN pin at one of the cascaded SAA4956TJs. A minimum delay of 40 SWCK clock cycles is needed before newly written data can be read back from memory (see Fig.16). If a reset read operation (RSTR) occurs in a read cycle before a reset write operation (RSTW) in a write cycle accessing the same location, then old data will be read. 7.1.7 TEST MODE OPERATION The SAA4956TJ incorporates a test mode not intended for customer use. If WE and RSTW are held HIGH continuously for 1024 SWCK clock cycles, the SAA4956TJ will enter test mode. It will exit test mode if WE is LOW for a single SWCK cycle or if RSTW is LOW for 2 SWCK clock cycles. Old data will be read provided a data read cycle begins within 20 pointer positions of the start of a write cycle. This means that if a reset read operation begins within 20 SWCK clock cycles after a reset write operation, the internal buffering of the SAA4956TJ will ensure that old data will be read out (see Fig.17). New data will be read if the read pointer is delayed by 40 pointer positions or more after the write pointer. Old data is still read out if the write pointer is less than or equal to 20 pointer positions ahead of the read pointer (internal buffering). A write pointer to read pointer delay of more than 20 but less than 40 pointer positions should be avoided. In this case, the old or the new data may be read, or a combination of both. In random read and write block access modes, the minimum write-to-read new data delay of 40 SWCK clock cycles must be inserted for each block. 1998 Dec 08 MEMORY ARBITRATION LOGIC AND SELF-REFRESH Since the data in the memory array is stored in DRAM cells, it needs to be refreshed periodically. Refresh is performed automatically under the control of internal memory arbitration logic which is clocked by a free running clock oscillator. The memory arbitration logic controls memory access for read, write and refresh operations. It uses the contents of the write, read and refresh address counters to access the memory array to load data from the parallel write register, store data in the parallel read register, or to refresh stored data. The values in these counters correspond to block addresses. If initialization starts earlier than the recommended 100 µs after power-up, the initialization sequence described above must be repeated, starting with an additional reset read operation and an additional reset write operation after the 100 µs start-up time. 7.1.4 SAA4956TJ 10 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 7.2 SAA4956TJ Noise reduction function 4 I2C-bus control: chroma_inverted DPCMin new Y old U/V + − delta U/V − LOW-PASS FILTER 1 I2C-bus control: unfiltered I2C-bus control: unfiltered − LF delta U/V + − LF delta Y U/V AVERAGE ABS/LIMITER LOW-PASS FILTER 2 LOW-PASS FILTER 2 I2C-bus control: chromafix and Klumatochroma Kchroma Kchromafix Kluma LUT + HF delta Y HF delta U/V ABS/LIMITER I2C-bus control: Yadapt_gain I2C-bus control: lumafix × Klumafix × Kluma LUT + × + + + processed U/V I2C-bus control: noise_shape NOISE SHAPE I2C-bus control: DPCMout FORMATTER processed Y I2C-bus control: noise_shape NOISE SHAPE 8 4 D to memory D0(V0), D1(V1) D2(U0), D3(U1) D to memory D11(Y7) to D4(Y0) MGR689 Switch position is off. Fig.3 Block diagram of noise reduction. 1998 Dec 08 old Y delta Y LOW-PASS FILTER 1 × 8 8 REFORMATTER new U/V I2C-bus control: Cadapt_gain D-field delay D11(Y7) to D4(Y0) 4 REFORMATTER + data input D11(Y7) to D4(Y0) D-field delay D0(V0), D1(V1) D2(U0), D3(U1) data input D0(V0), D1(V1) D2(U0), D3(U1) handbook, full pagewidth 11 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction The averaging in chrominance can be slaved optionally to the luminance averaging (I2C-bus control: Klumatochroma), therefore chrominance differences are not taken into account for the K-factor setting of the chrominance signal path. The main function of the noise reduction is shown in Fig.3. It is divided in two signal paths for chrominance and luminance. Two operating modes can be used in principal: the fixed and the adaptive mode. In both modes the applied frequency range, where the noise reduction takes place, can be reduced or not reduced (I2C-bus control: unfiltered). The noise reduction scheme effectively also decreases the cross-colour patterns if for the averaging in chrominance the adaptive noise reduction is slaved to the luminance averaging (I2C-bus control: Klumatochroma). The cross-colour pattern does not produce an increase of the measured luminance difference, therefore this pattern will be averaged over many fields. The noise reduction operates field recursive with an average ratio between fresh (new) and over previous fields averaged (old) luminance and chrominance. Noise reduction can be activated by forcing the NREN control signal to HIGH. In this case the system starts with default settings for noise reduction and noise shape, if clock is applied to SWCK. If NREN is LOW the noise reduction block is bridged via a data mux (see Fig.1). During NREN HIGH, only half of the write frequency, which is allowed during NREN LOW, can be applied. This is explained by the need for a second memory read access (READ2) to get old data (D-field delay) in the same frequency as the write frequency. The noise reduction is controlled via I2C-bus settings. Since the PAN-IC (SAA4995WP) needs approximately the same write frequency as the read frequency for the noise reduction option (32 MHz), a deactivated self-refresh is required. This setting can be done via the I2C-bus interface (I2C-bus control: no_refresh). 7.2.1 REFORMATTING AND FORMATTING The standard applications are using the 4 : 1 : 1 YUV data stream (see Table 1). The noise reduction processing uses internally the 4 : 2 : 2 data stream. An up converter to 4 : 2 : 2 is applied with a linear interpolation filter for creation of the extra samples. These are combined with the original samples from the 4 : 1 : 1 stream. Instead of the 4 : 1 : 1 mode the input or output colour can be handled as DPCM format which is a certain data compression (I2C-bus control: DPCMin, DPCMout) if the 4 : 2 : 2 data bandwidth is needed e.g. for applications together with SAA4978H. The incoming colour data can be also inverted for special purposes (I2C-bus control: Chroma_inverted). In the fixed mode, the noise reduction produces a constant weighted input averaging. Because of smearing effects this mode should not be used for normal operation except for K = 1. The fixed mode can be activated separately for chrominance (I2C-bus control: chromafix) and luminance (I2C-bus control: lumafix). Table 1 Digital input and output bus format INPUT PIN 4 : 1 : 1 FORMAT In the adaptive mode, the averaging ratio (K-factor) is based on the absolute differences of luminance respectively chrominance among the inputs. When the absolute difference is low, only a small part of the fresh data will be added. In cases of high difference, much of fresh data will be taken. This occurs in either situation of movement or where a significant vertical contrast is seen. The relation between the amount of movement and the K-factor values is defined via a Look-Up Table (LUT) where the steps can be programmed (I2C-bus controls: Ksteps). It should be noted that recursion is done over fields, and that pixel positions between the new and old fields always have a vertical offset of one line. So averaging is not only done in the dimension of time but also in the vertical direction. Therefore averaging vertically on e.g. a vertical black to a white edge would provide a grey result. 1998 Dec 08 SAA4956TJ 12 Y07 Y17 Y27 Y37 D11(Y7) Y06 Y16 Y26 Y36 D10(Y6) Y05 Y15 Y25 Y35 D9(Y5) Y04 Y14 Y24 Y34 D8(Y4) Y03 Y13 Y23 Y33 D7(Y3) Y02 Y12 Y22 Y32 D6(Y2) Y01 Y11 Y21 Y31 D5(Y1) Y00 Y10 Y20 Y30 D4(Y0) U07 U05 U03 U01 D3(U1) U06 U04 U02 U00 D2(U0) V07 V05 V03 V01 D1(V1) V06 V04 V02 V00 D0(V0) Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction The start position, when the first phase of the 4 : 1 : 1 YUV data word is expected on the input bus, is defined by the first valid written or read word after a reset write or a reset read operation. The luminance input signal is expected in 8-bit straight binary format, whereas U and V input signals are expected in twos complement format. U and V input signals are inverted if the corresponding I2C-bus control bit chroma_invert is set to logic 1. 7.2.2 The amplified signals, which correlate to the amount of movement in the chrominance respectively luminance signal path, are transferred into 1 out of 16 possible K-factor values via look-up tables. The K-factor values are defined by step values of the look-up tables. The step values are programmed via the I2C-bus-controls: Kstep0 to Kstep7 (see Table 3). These values are valid for the look-up tables of the chrominance and the luminance. For example, Kstep3 = 4 means: values which are smaller than 8 (4 times weight of 2) are at least K = 3⁄8 or smaller and values equal or higher than 8 are at least K = 4⁄8 or higher. BAND-SPLITTING After reformatting, the frequencies of the difference signals of luminance (delta Y) and chrominance (delta U/V) can be divided optionally in an upper band (HF) and lower band (LF) with a lowpass-filter in both signal paths. The lower frequency band signals (LF delta Y and LF delta U/V) will be used as input for the noise reduction function. The lower frequency band on these signals means that the specific picture contents does not move or is moving slowly. Optionally, it is possible to bridge the band-splitting (I2C-bus control: unfiltered = 1). 7.2.3 SAA4956TJ 7.2.4 K-FACTOR The amount of noise reduction (field averaging) is described via the K-factor. K = 1 means that no averaging is applied and the new field information is used. K = 0 means that no averaging is applied and thus only the old field information is used like a still picture mode. All values in between mean that a weighted averaging is applied. It is possible to use fixed K-factor values if the I2C-bus variable lumafix respectively chromafix is set to logic 1. The possible fixed K-factor values of the I2C-bus control variables Klumafix and Kchromafix are described in Table 3. MOTION DETECTION The same signals, on which the noise reduction will be applied, are used to detect the amount of motion in the difference signals. Therefore, the absolute value of the difference signals are generated and limited to a maximum value. Afterwards the absolute values of the difference signal of U and V are averaged. The signals are low-pass filtered to smooth these signals. The filtered signals are amplified, depending on the setting of the I2C-bus controls: Yadapt_gain respectively Cadapt_gain (see Table 2). Table 3 Settings of fixed K-factor values Klumafix/Kchromafix[3 to 0] K-factor HEX DECIMAL 00 00 01 16 02 02 2⁄ 16 03 03 3⁄ 16 04 04 4⁄ 16 05 05 5⁄ 16 06 06 6⁄ 16 07 07 7⁄ 16 08 08 8⁄ 16 09 09 9⁄ 0A 10 01 Table 2 Gain settings of adaptive values for chrominance and luminance Yadapt_gain/Cadapt_gain[2 to 0] GAIN HEX DECIMAL 00 00 01 01 02 02 03 03 04 04 05 05 06 06 07 07 1998 Dec 08 1⁄ 8 2⁄ 8 4⁄ 8 8⁄ 8 16⁄ 8 32⁄ 8 64⁄ 8 128⁄ 8 13 0 1⁄ 0B 11 0C 12 0D 13 0E 14 0F 15 16 10⁄ 16 11⁄ 16 12⁄ 16 13⁄ 16 14⁄ 16 16⁄ 16 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 7.2.5 There is no ‘wrap around’ of subaddresses. Commands and data are processed as soon as they have been completely received. Data patterns sent to the various subaddresses are not checked for being illegal or not at that address. Detection of a STOP condition without a preceding acknowledge bit is regarded as a bus error. The last operation will not then be executed. Invalid subaddresses are not acknowledged. NOISE SHAPE If the noise shaping is activated possible shadow picture information in the chrominance and the luminance path, resulting from a low K-factor value, will be eliminated. The noise shaping function can be switched off by means of the I2C-bus control. Subregister noise_shape = 0 is applied to show the effect. The default I2C-bus settings can be loaded by changing the state of the NREN pin from LOW to HIGH during clocking of the SWCK and SCL pins. This can be realized for example by delaying the NREN signal with an RC-circuit for 1 second during the power-up sequence. This is necessary to give the circuit a minimum of 20 cycles of SCL and 100 cycles of SWCK after all input signals are stabilized. I2C-bus interface 7.3 SAA4956TJ The I2C-bus interface in the SAA4956TJ is used in a receive mode. The standardized bus frequencies of both 100 kHz and 400 kHz can be dealt with. As a slave receiver, the SAA4956TJ provides 8 registers for storing commands and data. These registers are accessed via so-called subaddresses. A subaddress can be thought of as a pointer to an internal memory location. It is allowed to send one data byte or more data bytes per transmission to the SAA4956TJ. In this event, the subaddress is automatically incremented after each data byte, resulting in storing the sequence of data bytes at successive register locations, starting at SUBADDRESS. A transmission can start at any valid subaddress. Each data byte is acknowledged with ACK (acknowledge). Table 4 S I2C-bus control; slave address, subaddress and data format SLAVE ADDRESS Table 5 0 ACK SUBADDRESS ACK DATA Description of bits used in Table 4 BIT FUNCTION S START condition SLAVE ADDRESS 7-bit device address: 1011 111 (last bit is LSB) 0 data direction bit (write to device) ACK acknowledge SUBADDRESS address of register to write to DATA data byte to be written into register P STOP condition 1998 Dec 08 14 ACK ... DATA ACK P Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction Table 6 SAA4956TJ Write registers REGISTER BIT NAME DEFAULT FUNCTION Subregister 20H to 23H (Ksteps of the look-up tables of luminance and chrominance) 20H 21H 22H 23H 0 to 3 Kstep0 2 step in adaptive curve from K = 1⁄16 to K = 1⁄8; weight of 1 4 to 7 Kstep1 3 step in adaptive curve from K = 1⁄8 to K = 2⁄8; weight of 1 0 to 3 Kstep2 4 step in adaptive curve from K = 2⁄8 to K = 3⁄8; weight of 2 4 to 7 Kstep3 6 step in adaptive curve from K = 3⁄8 to K = 4⁄8; weight of 2 0 to 3 Kstep4 4 step in adaptive curve from K = 4⁄8 to K = 5⁄8; weight of 4 4 to 7 Kstep5 6 step in adaptive curve from K = 5⁄8 to K = 6⁄8; weight of 4 0 to 3 Kstep6 4 step in adaptive curve from K = 6⁄8 to K = 7⁄8; weight of 8 4 to 7 Kstep7 6 step in adaptive curve from K = 7⁄8 to K = 8⁄8; weight of 8 Subregister 24H (K-factor controls of luminance) 24H 0 to 3 Klumafix 15 value of the fixed K-factor of the luminance; see Table 3 4 to 6 Yadapt_gain 1 value of the gain of the adaptive curve of the luminance; see Table 2 7 0 adaptive (lumafix = 0) or fixed K mode (lumafix = 1) of the luminance lumafix Subregister 25H (K-factor controls of chrominance) 25H 0 to 3 Kchromafix 15 value of the fixed K-factor of the luminance; see Table 3 4 to 6 Cadapt_gain 1 value of the gain of the adaptive curve of the chrominance; see Table 2 7 0 adaptive (chromafix = 0) or fixed K mode (chromafix = 1) of chrominance chromafix Subregister 26H (miscellaneous) 26H 0 DPCMin 0 if HIGH: converts the chrominance input from DPCM to 4 : 2 : 2 format 1 DPCMout 0 if HIGH: converts the chrominance output from 4 : 2 : 2 to DPCM format 2 Klumatochroma 0 if HIGH: uses luminance K-factor for chrominance path 3 unfiltered 0 if HIGH: band splitting is deactivated, complete difference signals are used 4 noise_shape 1 if HIGH: noise shaping is activated 5 chroma_invert 0 if HIGH: chrominance input signals U and V are inverted 6 split_screen 0 if HIGH: demo mode is activated with noise reduction only at the right screen side 7 reserved Subregister 27H (miscellaneous) 27H 0 no_refresh 0 if HIGH: disables internal self-refresh to allow Tcy(SWCK) = 29 ns for PAN-IC 1 field_sync 0 if HIGH: synchronizes the noise reduction via RSTW (split_screen and DPCM not possible). Otherwise sync is line related detected by WE 5 cycle LOW followed by a HIGH cycle. 2 to 7 reserved 1998 Dec 08 15 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ 8 LIMITING VALUES In accordance with the Absolute Maximum Rating System (IEC 134). SYMBOL PARAMETER CONDITIONS MIN. MAX. UNIT VDD supply voltage (pin 19) −0.5 +5 V VDD(O) supply voltage (pin 22) −0.5 +5 V VDD(P) supply voltage for protection circuits −0.5 +5.5 V VI input voltage −0.5 +5.5 V VDD = VDD(O) = VDD(P) = 3.3 V −0.5 +3.8 V +5 V +3.8 V VO VDD(P) = 5 V output voltage −0.5 VDD(P) = 5 V VDD = VDD(O) = VDD(P) = 3.3 V −0.5 − 200 mA ∆VGND-OGND voltage difference between GND and OGND −0.5 +0.5 V IO(sc) short-circuit output current − 50 mA Ptot total power dissipation − 750 mW Tstg storage temperature −20 +150 °C Tj junction temperature 0 125 °C Tamb ambient temperature Ves electrostatic handling IDD(tot) total supply current 0 70 °C note 1 −200 +200 V note 2 −2000 +2000 V Notes 1. Machine model: equivalent to discharging a 200 pF capacitor through a 0 Ω series resistor (‘0 Ω’ is actually 0.75 µH + 10 Ω). 2. Human body model: equivalent to discharging a 100 pF capacitor through a 1500 Ω series resistor. 9 THERMAL CHARACTERISTICS SYMBOL Rth(j-a) 1998 Dec 08 PARAMETER CONDITIONS thermal resistance from junction to ambient 16 in free air VALUE UNIT 60 K/W Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ 10 CHARACTERISTICS VDD = VDD(O) = VDD(P) = 3.0 to 3.6 V; Tamb = 0 to 70 °C; 3 ns input transition times; unless otherwise specified. SYMBOL PARAMETER CONDITIONS MIN. TYP.(1) MAX. UNIT Supply VDD supply voltage (pin 19) 3.0 3.3 3.6 V VDD(O) supply voltage (pin 22) 3.0 3.3 3.6 V VDD(P) supply voltage (pin 21) 3.0 3.3 5.5 V IDD(tot) total supply current (IDD(tot) = IDD + IDD(O) + IDD(P)) minimum write/read cycle; outputs open-circuit − 27 70 mA IDD operating supply current minimum write/read cycle − 25 60 mA IDD(std) stand-by supply current after 1 RSTW/RSTR cycle; NREN, WE, RE and OE = LOW − 3 10 mA IDD(O) supply current minimum write/read cycle; outputs open-circuit − 2 10 mA IDD(P) supply current − 0 1 mA Inputs except I2C-bus signals (pins 3 to 18, 23 to 26 and 40) VIH HIGH-level input voltage 2.0 − VDD(P) + 0.3 V VIL LOW-level input voltage −0.5 − +0.8 V ILI input leakage current Vi = 0 V to VDD(P) −10 − +10 µA Ci input capacitance f = 1 MHz; Vi = 0 V − − 7 pF Inputs of I2C-bus signals: SCL (pin 1) and SDA (pin 20); note 2 VIH HIGH-level input voltage 3.0 − VDD(P) + 0.5 V VIL LOW-level input voltage −0.5 − +1.5 ILI input leakage current Vi = 0 V to VDD(P) −10 − +10 µA Ci input capacitance f = 1 MHz; Vi = 0 V − − 10 pF 2.4 − − V V Outputs except I2C-bus signal SCL (pins 27 to 38) VOH HIGH-level output voltage IOH = −5 mA VOL LOW-level output voltage IOL = 4.2 mA ILO output leakage current Co output capacitance f = 1 MHz; Vo = 0 V − − 0.4 V −10 − +10 µA − − 10 pF Output of I2C-bus signal: SDA (pin 20); note 2 VOL LOW-level output voltage ILO output leakage current Co output capacitance − − 0.4 V −10 − +10 µA f = 1 MHz; Vo = 0 V − − 10 pF NREN = LOW; see Fig.4 26 − − ns NREN = HIGH; see Fig.4 52 − 150 ns IOL = 4 mA Write cycle timing; note 3 Tcy(SWCK) SWCK cycle time tW(SWCKH) SWCK HIGH pulse width see Fig.4 7 − − ns tW(SWCKL) SWCK LOW pulse width see Fig.4 7 − − ns 1998 Dec 08 17 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SYMBOL PARAMETER CONDITIONS SAA4956TJ MIN. TYP.(1) MAX. UNIT tsu(D) set-up time data inputs (D0 to D11) see Fig.4 5 − − ns th(D) hold time data inputs (D0 to D11) see Fig.4 3 − − ns tsu(RSTW) set-up time RSTW see Fig.4 5 − − ns th(RSTW) hold time RSTW see Fig.4 3 − − ns tsu(WE) set-up time WE see Fig.8 5 − − ns th(WE) hold time WE see Fig.8 3 − − ns tW(WEL) WE LOW pulse width see Fig.8 8 − − ns tsu(IE) set-up time IE see Fig.9 5 − − ns th(IE) hold time IE see Fig.9 3 − − ns tW(IEL) IE LOW pulse width see Fig.9 8 − − ns tt transition time (rise and fall) see Fig.4 − 3 30 ns Read cycle timing; note 4 tACC access time after SRCK see Fig.11 − − 21 ns ten(Q) output enable time after SRCK see Fig.15 − − 21 ns tdis(Q) output disable time after SRCK note 5; see Fig.15 − − 12 ns th(Q) output hold time after SRCK see Fig.11 3 − − ns Tcy(SRCK) SRCK cycle time see Fig.11 26 − − ns tW(SRCKH) HIGH-level pulse width of SRCK see Fig.11 7 − − ns tW(SRCKL) LOW-level pulse width of SRCK see Fig.11 7 − − ns tsu(RSTR) set-up time RSTR see Fig.11 5 − − ns th(RSTR) hold time RSTR see Fig.11 3 − − ns tsu(RE) set-up time RE see Fig.14 5 − − ns th(RE) hold time RE see Fig.14 3 − − ns tW(REL) LOW-level pulse width of RE see Fig.14 9 − − ns tsu(OE) set-up time OE see Fig.15 5 − − ns th(OE) hold time OE see Fig.15 3 − − ns tW(OEL) LOW-level pulse width of OE see Fig.15 9 − − ns tt transition time (rise and fall) see Fig.11 − 3 30 ns Notes 1. Typical values are valid for Tamb = 25 °C, VDD = VDD(O) = VDD(P) = 3.3 V, all voltages referenced to GND. See Fig.1 for configuration. 2. The AC characteristics are in accordance with the I2C-bus specification for fast mode (clock frequency maximum 400 kHz). Information about the I2C-bus can be found in the brochure “The I2C-bus and how to use it” (order number 9398 393 40011). 3. The write cycle timing set-up and hold times are related to VIL of the rising edge of SWCK. They are valid for the specified LOW and HIGH-level input voltages (VIL and VIH). 4. The read cycle timing set-up and hold times are related to VIL of the rising edge of SRCK. They are valid for the specified LOW and HIGH-level input voltages (VIL and VIH). The load on each output is a 30 pF capacitor to ground in parallel with a 218 Ω resistor to 1.31 V. 5. Disable times specified are from the initiating edge until the output is no longer driven by the memory. Disable times are measured by observing the output waveforms. Low values of load resistor and capacitor have to be used to obtain a short time constant. 1998 Dec 08 18 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction N−2 handbook, full pagewidth N−1 SAA4956TJ 1 N 2 Tcy(SWCK) −VIH SWCK −VIL th(RSTW) tw(SWCKH) tw(SWCKL) th(RSTW) tt tsu(RSTW) tsu(RSTW) −VIH RSTW −VIL tsu(D) th(D) D0 to D11 N−2 N−1 N 1 −VIH 2 −VIL −VIH WE −VIL −VIH IE −VIL MGK677 Fig.4 Write cycle timing diagram (reset write). handbook, full pagewidth N−1 N disable disable 1 Tcy(SWCK) −VIH SWCK −VIL tw(SWCKH) tw(SWCKL) −VIH RSTW D0 to D11 −VIL N−1 N 1 −VIL −VIH WE −VIL −VIH IE −VIL MGK678 Fig.5 Write cycle timing diagram (reset write with WE LOW). 1998 Dec 08 −VIH 19 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction handbook, full pagewidth 1 start sequence (4 SWCK) 2 3 SAA4956TJ serial input of write block address (13 SWCK) 4 5 write latency (minimum 18 SWCK) 17 18 19 34 write data 35 36 SWCK RSTW WE LOW: D0 controlled WE IE random write block address D0 MSB write data 0 LSB 1 2 3 at block address: write data 0 D1 to D11 1 2 3 MGK679 Fig.6 D0 controlled entry sequence of the random write block access mode. handbook, full pagewidth 1 start sequence (4 SWCK) 2 3 serial input of write block address (13 SWCK) 4 5 write latency (minimum 18 SWCK) 17 18 19 34 write data 35 36 SWCK RSTW WE HIGH: IE controlled WE random write block address IE MSB LSB at block address: write data 0 D0 to D11 1 2 3 MGK680 Fig.7 IE controlled entry sequence of the random write block access mode. 1998 Dec 08 20 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction N−1 handbook, full pagewidth disable N SAA4956TJ N+1 disable −VIH SWCK −VIL th(WE) tsu(WE) th(WE) tsu(WE) −VIH WE −VIL tw(WEL) tsu(D) D0 to D11 N−1 th(D) N+1 N −VIH −VIL −VIH IE −VIL −VIH RSTW −VIL MGK681 Fig.8 Write cycle timing diagram (write enable). N−1 handbook, full pagewidth disable N N+3 disable −VIH SWCK −VIL th(IE) tsu(IE) th(IE) tsu(IE) −VIH IE −VIL tw(IEL) tsu(D) D0 to D11 N−1 th(D) N+3 N −VIL −VIH WE −VIL −VIH RSTW −VIL MGK682 Fig.9 Write cycle timing diagram (input enable = write mask operation). 1998 Dec 08 −VIH 21 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction N handbook, full pagewidth N+1 N+2 N+3 SAA4956TJ N+4 N+5 N+6 N+7 −VIH SWCK −VIL −VIH IE −VIL −VIH WE −VIL D0 to D11 N N+1 N+2 N+3 N+6 N+7 −VIH N+8 −VIL −VIH RSTW −VIL N N+1 N+2 N+3 N+4 N+5 N+6 N+7 N+8 −VIH SRCK −VIL −VIH OE −VIL −VIH RE −VIL new Q0 to Q11 high-Z N new old old new new new N+3 N+4 N+5 N+6 N+7 N+8 −VIL −VIH RSTR −VIL MGK683 Fig.10 Write mask operation. 1998 Dec 08 −VIH 22 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction N−1 handbook, full pagewidth 1 N SAA4956TJ 2 3 Tcy(SRCK) −VIH SRCK −VIL th(RSTR) tw(SRCKH) tw(SRCKL) th(RSTR) tsu(RSTR) tt tsu(RSTR) −VIH RSTR −VIL th(Q) tACC Q0 to Q11 N−2 N−1 N 1 2 −VIH −VIL −VIH RE −VIL −VIH OE −VIL MGK684 Fig.11 Read cycle timing diagram (reset read). handbook, full pagewidthN N N 1 2 Tcy(SRCK) −VIH SRCK −VIL tw(SRCKH) −VIH tw(SRCKL) RSTR −VIL tACC Q0 to Q11 N−1 tACC N −VIH 1 −VIL −VIH RE −VIL −VIH OE −VIL MGK685 Fig.12 Read cycle timing diagram (reset read with RE LOW). 1998 Dec 08 23 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction handbook, full pagewidth 1 start sequence (4 SRCK) 2 3 SAA4956TJ serial input of read block address (13 SRCK) 4 5 read latency (minimum 20 SRCK) 17 18 19 36 37 read data 38 SRCK RSTR RE random read block address MSB OE LSB at block address: read data high-Z Q0 to Q11 0 1 2 MGK686 Fig.13 Entry sequence of the random read block access mode. handbook, full pagewidthN N+1 N N N+2 −VIH SRCK −VIL th(RE) tsu(RE) th(RE) tsu(RE) −VIH RE −VIL tw(REL) Q0 to Q11 N−1 tACC N N+1 −VIL −VIH OE −VIL −VIH RSTR −VIL MGK687 Fig.14 Read cycle timing diagram (read enable). 1998 Dec 08 −VIH 24 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction handbook, full pagewidth N disable SAA4956TJ N+3 disable N+4 −VIH SRCK −VIL th(OE) tsu(OE) th(OE) tsu(OE) −VIH OE −VIL tw(OEL) tACC ten(Q) tdis(Q) Q0 to Q11 N−1 high-Z N N+3 −VIL −VIH RE −VIL −VIH RSTR −VIL MGK688 Fig.15 Read cycle timing diagram (output enable). 1998 Dec 08 −VIH 25 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 1 handbook, full pagewidth 2 3 SAA4956TJ 39 40 41 42 −VIH SWCK −VIL −VIH RSTW −VIL −VIH WE and IE −VIL D0 to D11 new new new new new new new 1 2 3 39 40 41 42 −VIH −VIL minimum number of SWCK cycles delay to get new data 1 2 3 −VIH SRCK −VIL −VIH RSTR −VIL −VIH RE and OE Q1 to Q11 −VIL new new 1 2 −VIH high-Z −VIL MGK689 Fig.16 New data access. 1998 Dec 08 26 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 1 handbook, full pagewidth 2 3 SAA4956TJ 19 20 21 22 −VIH SWCK −VIL −VIH RSTW −VIL −VIH WE and IE −VIL D0 to D11 new new new new new new new 1 2 3 19 20 21 22 −VIH −VIL maximum number of SWCK cycles delay to get old data 1 2 3 −VIH SRCK −VIL −VIH RSTR −VIL −VIH RE and OE Q1 to Q11 −VIL old old 1 2 −VIH high-Z −VIL MGK690 Fig.17 Old data access. 1998 Dec 08 27 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ reset handbook, full pagewidth signal serial clock RSTW SWCK data inputs 16 25 15 26 D0(V0) to D11(Y7) RSTW SRCK SWCK Q0(V0) to Q11(Y7) SAA4956TJ 14 to 3 RSTR 16 25 15 26 D0(V0) to D11(Y7) 27 to 38 SRCK Q0(V0) to Q11(Y7) SAA4956TJ 14 to 3 12 RSTR 12 WE IE 17 24 18 23 data outputs 27 to 38 12 RE WE OE IE 17 24 18 23 RE OE enable signal MGR690 Fig.18 Cascade operation (signal connections). write new data read 2 times delayed old data handbook, full pagewidth 1 1 2 2 3 3 4 4 5 5 6 6 7 SWCK and SRCK −VIH RSTW and RSTR −VIH WE and IE and RE and OE −VIH −VIL −VIL −VIL data inputs (×12) data outputs (×12) new new new new new new 1 2 3 4 5 6 old old old old old old 1 2 3 4 5 6 −VIH −VIL −VIH high-Z −VIL MGK692 Fig.19 Cascade operation (timing waveforms). 1998 Dec 08 28 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ 11 APPLICATION INFORMATION handbook, full pagewidth +3.3 V YIN 26 UIN 28 VIN D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 +5 V SDA 10 kΩ NREN 4.7 µF 17, 18, 19, 23, 25, 29, RST 46, 67 9 30 +3.3 V SCL 8, 11, 69, 75, 80 +5 V 19, 22 +5 V 21, 23 1 15 3 16 SWC RSTW WE 47 45 24 44 32 43 4 17, 18 5 38 51 42 6 37 52 41 7 36 53 40 8 35 54 39 34 55 10 33 56 37 11 32 57 36 12 31 58 35 13 30 59 34 14 29 60 20 28 61 25 27 26 24 9 40 SAA4956TJ SAA4977H 62 RE 38 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 79 YOUT 76 UOUT 74 VOUT 63 2, 39 1 SDA 2 SCL 20 22 HRD HDFL 71 14 to 16, 21, 27, 31, 72 48 to 50, 3 to 7, 68 33, 65, 73, 10, 12, 70 77, 78 13, 64, 66 VDFL n.c. VA DISPLAY PLL HA SRC MBK914 Fig.20 Application diagram. 1998 Dec 08 29 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ 12 PACKAGE OUTLINE SOJ40: plastic small outline package; 40 leads (J-bent); body width 10.16 mm SOT449-1 X D c eE y bp b1 40 A 21 w M E HE A2 A pin 1 index A1 (A 3) 1 20 Lp ZD e detail X v M A 0 5 10 mm scale DIMENSIONS (millimetre dimensions are derived from the original inch dimensions) UNIT A max. A1 A2 A3 bp b1 c D(1) E(1) e eE HE Lp v w y mm 3.68 1.40 1.14 2.29 2.18 0.25 0.51 0.38 0.81 0.66 0.32 0.18 26.2 25.9 10.3 10.0 1.27 9.4 11.30 11.05 1.4 1.1 0.18 0.18 0.1 ZD (1) 1.19 0.73 Note 1. Plastic or metal protrusions of 0.25 mm maximum per side are not included. OUTLINE VERSION SOT449-1 1998 Dec 08 REFERENCES IEC JEDEC EIAJ EUROPEAN PROJECTION ISSUE DATE 97-06-02 MS027 30 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction • Use a double-wave soldering method comprising a turbulent wave with high upward pressure followed by a smooth laminar wave. 13 SOLDERING 13.1 Introduction to soldering surface mount packages • For packages with leads on two sides and a pitch (e): This text gives a very brief insight to a complex technology. A more in-depth account of soldering ICs can be found in our “Data Handbook IC26; Integrated Circuit Packages” (document order number 9398 652 90011). – larger than or equal to 1.27 mm, the footprint longitudinal axis is preferred to be parallel to the transport direction of the printed-circuit board; – smaller than 1.27 mm, the footprint longitudinal axis must be parallel to the transport direction of the printed-circuit board. There is no soldering method that is ideal for all surface mount IC packages. Wave soldering is not always suitable for surface mount ICs, or for printed-circuit boards with high population densities. In these situations reflow soldering is often used. 13.2 The footprint must incorporate solder thieves at the downstream end. • For packages with leads on four sides, the footprint must be placed at a 45° angle to the transport direction of the printed-circuit board. The footprint must incorporate solder thieves downstream and at the side corners. Reflow soldering 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. 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. Several methods exist for reflowing; for example, infrared/convection heating in a conveyor type oven. Throughput times (preheating, soldering and cooling) vary between 100 and 200 seconds depending on heating method. 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. Typical reflow peak temperatures range from 215 to 250 °C. The top-surface temperature of the packages should preferable be kept below 230 °C. 13.3 13.4 Manual soldering Fix the component by first soldering two diagonally-opposite end leads. Use a low voltage (24 V or less) soldering iron applied to the flat part of the lead. Contact time must be limited to 10 seconds at up to 300 °C. Wave soldering Conventional single wave soldering is not recommended for surface mount devices (SMDs) or printed-circuit boards with a high component density, as solder bridging and non-wetting can present major problems. When using a dedicated tool, all other leads can be soldered in one operation within 2 to 5 seconds between 270 and 320 °C. To overcome these problems the double-wave soldering method was specifically developed. If wave soldering is used the following conditions must be observed for optimal results: 1998 Dec 08 SAA4956TJ 31 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction 13.5 SAA4956TJ Suitability of surface mount IC packages for wave and reflow soldering methods SOLDERING METHOD PACKAGE REFLOW(1) WAVE BGA, SQFP not suitable HLQFP, HSQFP, HSOP, SMS not PLCC(3), SO, SOJ suitable suitable(2) suitable suitable suitable LQFP, QFP, TQFP not recommended(3)(4) suitable SSOP, TSSOP, VSO not recommended(5) suitable Notes 1. All surface mount (SMD) packages are moisture sensitive. Depending upon the moisture content, the maximum temperature (with respect to time) and body size of the package, there is a risk that internal or external package cracks may occur due to vaporization of the moisture in them (the so called popcorn effect). For details, refer to the Drypack information in the “Data Handbook IC26; Integrated Circuit Packages; Section: Packing Methods”. 2. These packages are not suitable for wave soldering as a solder joint between the printed-circuit board and heatsink (at bottom version) can not be achieved, and as solder may stick to the heatsink (on top version). 3. If wave soldering is considered, then the package must be placed at a 45° angle to the solder wave direction. The package footprint must incorporate solder thieves downstream and at the side corners. 4. Wave soldering is only suitable for LQFP, TQFP and QFP packages with a pitch (e) equal to or larger than 0.8 mm; it is definitely not suitable for packages with a pitch (e) equal to or smaller than 0.65 mm. 5. Wave soldering is only suitable for SSOP and TSSOP packages with a pitch (e) equal to or larger than 0.65 mm; it is definitely not suitable for packages with a pitch (e) equal to or smaller than 0.5 mm. 1998 Dec 08 32 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction SAA4956TJ 14 DEFINITIONS Data sheet status Objective specification This data sheet contains target or goal specifications for product development. Preliminary specification This data sheet contains preliminary data; supplementary data may be published later. Product specification 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. 15 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. 16 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. 1998 Dec 08 33 Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction NOTES 1998 Dec 08 34 SAA4956TJ Philips Semiconductors Preliminary specification 2.9-Mbit field memory with noise reduction NOTES 1998 Dec 08 35 SAA4956TJ Philips Semiconductors – a worldwide company Argentina: see South America Australia: 34 Waterloo Road, NORTH RYDE, NSW 2113, Tel. +61 2 9805 4455, Fax. +61 2 9805 4466 Austria: Computerstr. 6, A-1101 WIEN, P.O. Box 213, Tel. +43 160 1010, Fax. +43 160 101 1210 Belarus: Hotel Minsk Business Center, Bld. 3, r. 1211, Volodarski Str. 6, 220050 MINSK, Tel. +375 172 200 733, Fax. +375 172 200 773 Belgium: see The Netherlands Brazil: see South America Bulgaria: Philips Bulgaria Ltd., Energoproject, 15th floor, 51 James Bourchier Blvd., 1407 SOFIA, Tel. +359 2 689 211, Fax. +359 2 689 102 Canada: PHILIPS SEMICONDUCTORS/COMPONENTS, Tel. +1 800 234 7381 China/Hong Kong: 501 Hong Kong Industrial Technology Centre, 72 Tat Chee Avenue, Kowloon Tong, HONG KONG, Tel. +852 2319 7888, Fax. +852 2319 7700 Colombia: see South America Czech Republic: see Austria Denmark: Prags Boulevard 80, PB 1919, DK-2300 COPENHAGEN S, Tel. +45 32 88 2636, Fax. +45 31 57 0044 Finland: Sinikalliontie 3, FIN-02630 ESPOO, Tel. +358 9 615800, Fax. +358 9 61580920 France: 51 Rue Carnot, BP317, 92156 SURESNES Cedex, Tel. +33 1 40 99 6161, Fax. +33 1 40 99 6427 Germany: Hammerbrookstraße 69, D-20097 HAMBURG, Tel. +49 40 23 53 60, Fax. +49 40 23 536 300 Greece: No. 15, 25th March Street, GR 17778 TAVROS/ATHENS, Tel. +30 1 4894 339/239, Fax. +30 1 4814 240 Hungary: see Austria India: Philips INDIA Ltd, Band Box Building, 2nd floor, 254-D, Dr. Annie Besant Road, Worli, MUMBAI 400 025, Tel. +91 22 493 8541, Fax. +91 22 493 0966 Indonesia: PT Philips Development Corporation, Semiconductors Division, Gedung Philips, Jl. Buncit Raya Kav.99-100, JAKARTA 12510, Tel. +62 21 794 0040 ext. 2501, Fax. +62 21 794 0080 Ireland: Newstead, Clonskeagh, DUBLIN 14, Tel. +353 1 7640 000, Fax. +353 1 7640 200 Israel: RAPAC Electronics, 7 Kehilat Saloniki St, PO Box 18053, TEL AVIV 61180, Tel. +972 3 645 0444, Fax. +972 3 649 1007 Italy: PHILIPS SEMICONDUCTORS, Piazza IV Novembre 3, 20124 MILANO, Tel. +39 2 6752 2531, Fax. +39 2 6752 2557 Japan: Philips Bldg 13-37, Kohnan 2-chome, Minato-ku, TOKYO 108-8507, Tel. +81 3 3740 5130, Fax. +81 3 3740 5077 Korea: Philips House, 260-199 Itaewon-dong, Yongsan-ku, SEOUL, Tel. +82 2 709 1412, Fax. +82 2 709 1415 Malaysia: No. 76 Jalan Universiti, 46200 PETALING JAYA, SELANGOR, Tel. +60 3 750 5214, Fax. +60 3 757 4880 Mexico: 5900 Gateway East, Suite 200, EL PASO, TEXAS 79905, Tel. +9-5 800 234 7381 Middle East: see Italy Netherlands: Postbus 90050, 5600 PB EINDHOVEN, Bldg. VB, Tel. +31 40 27 82785, Fax. +31 40 27 88399 New Zealand: 2 Wagener Place, C.P.O. Box 1041, AUCKLAND, Tel. +64 9 849 4160, Fax. +64 9 849 7811 Norway: Box 1, Manglerud 0612, OSLO, Tel. +47 22 74 8000, Fax. +47 22 74 8341 Pakistan: see Singapore Philippines: Philips Semiconductors Philippines Inc., 106 Valero St. Salcedo Village, P.O. Box 2108 MCC, MAKATI, Metro MANILA, Tel. +63 2 816 6380, Fax. +63 2 817 3474 Poland: Ul. Lukiska 10, PL 04-123 WARSZAWA, Tel. +48 22 612 2831, Fax. +48 22 612 2327 Portugal: see Spain Romania: see Italy Russia: Philips Russia, Ul. Usatcheva 35A, 119048 MOSCOW, Tel. +7 095 755 6918, Fax. +7 095 755 6919 Singapore: Lorong 1, Toa Payoh, SINGAPORE 319762, Tel. +65 350 2538, Fax. +65 251 6500 Slovakia: see Austria Slovenia: see Italy South Africa: S.A. PHILIPS Pty Ltd., 195-215 Main Road Martindale, 2092 JOHANNESBURG, P.O. Box 7430 Johannesburg 2000, Tel. +27 11 470 5911, Fax. +27 11 470 5494 South America: Al. Vicente Pinzon, 173, 6th floor, 04547-130 SÃO PAULO, SP, Brazil, Tel. +55 11 821 2333, Fax. +55 11 821 2382 Spain: Balmes 22, 08007 BARCELONA, Tel. +34 93 301 6312, Fax. +34 93 301 4107 Sweden: Kottbygatan 7, Akalla, S-16485 STOCKHOLM, Tel. +46 8 5985 2000, Fax. +46 8 5985 2745 Switzerland: Allmendstrasse 140, CH-8027 ZÜRICH, Tel. +41 1 488 2741 Fax. +41 1 488 3263 Taiwan: Philips Semiconductors, 6F, No. 96, Chien Kuo N. Rd., Sec. 1, TAIPEI, Taiwan Tel. +886 2 2134 2865, Fax. +886 2 2134 2874 Thailand: PHILIPS ELECTRONICS (THAILAND) Ltd., 209/2 Sanpavuth-Bangna Road Prakanong, BANGKOK 10260, Tel. +66 2 745 4090, Fax. +66 2 398 0793 Turkey: Talatpasa Cad. No. 5, 80640 GÜLTEPE/ISTANBUL, Tel. +90 212 279 2770, Fax. +90 212 282 6707 Ukraine: PHILIPS UKRAINE, 4 Patrice Lumumba str., Building B, Floor 7, 252042 KIEV, Tel. +380 44 264 2776, Fax. +380 44 268 0461 United Kingdom: Philips Semiconductors Ltd., 276 Bath Road, Hayes, MIDDLESEX UB3 5BX, Tel. +44 181 730 5000, Fax. +44 181 754 8421 United States: 811 East Arques Avenue, SUNNYVALE, CA 94088-3409, Tel. +1 800 234 7381 Uruguay: see South America Vietnam: see Singapore Yugoslavia: PHILIPS, Trg N. Pasica 5/v, 11000 BEOGRAD, Tel. +381 11 625 344, Fax.+381 11 635 777 For all other countries apply to: Philips Semiconductors, International Marketing & Sales Communications, Building BE-p, P.O. Box 218, 5600 MD EINDHOVEN, The Netherlands, Fax. +31 40 27 24825 Internet: http://www.semiconductors.philips.com © Philips Electronics N.V. 1998 SCA60 All rights are reserved. Reproduction in whole or in part is prohibited without the prior written consent of the copyright owner. The information presented in this document does not form part of any quotation or contract, is believed to be accurate and reliable and may be changed without notice. No liability will be accepted by the publisher for any consequence of its use. Publication thereof does not convey nor imply any license under patent- or other industrial or intellectual property rights. Printed in The Netherlands 545104/00/01/pp36 Date of release: 1998 Dec 08 Document order number: 9397 750 04286