CY14C256I CY14B256I, CY14E256I 2 256-Kbit (32 K × 8) Serial (I C) nvSRAM with Real Time Clock 256-Kbit (32 K × 8) Serial (I2C) nvSRAM with Real Time Clock Features ■ 256-Kbit nonvolatile static random access memory (nvSRAM) ❐ Internally organized as 32 K × 8 ❐ STORE to QuantumTrap nonvolatile elements initiated automatically on power-down (AutoStore) or by using I2C command (Software STORE) or HSB pin (Hardware STORE) ❐ RECALL to SRAM initiated on power-up (Power-Up RECALL) or by I2C command (Software RECALL) ❐ Automatic STORE on power-down with a small capacitor ■ High reliability ■ Infinite read, write, and RECALL cycles 1 million STORE cycles to QuantumTrap ❐ Data retention: 20 years at 85 °C ■ Real Time Clock (RTC) ❐ Full-featured RTC ❐ Watchdog timer ❐ Clock alarm with programmable interrupts ❐ Backup power fail indication ❐ Square wave output with programmable frequency (1 Hz, 512 Hz, 4096 Hz, 32.768 kHz) ❐ Capacitor or battery backup for RTC ❐ Backup current of 0.45 µA (typical) 2 ■ High-speed I C interface ❐ Industry standard 100 kHz and 400 kHz speed ❐ Fast mode Plus 1 MHz speed ❐ High speed: 3.4 MHz ❐ Zero cycle delay reads and writes ■ Write protection ❐ Hardware protection using Write Protect (WP) pin ❐ Software block protection for one-quarter, one-half, or entire array ■ ■ ❐ ❐ Logic Block Diagram I2C access to special functions ❐ Nonvolatile STORE/RECALL ❐ 8-byte serial number ❐ Manufacturer ID and Product ID ❐ Sleep mode Low power consumption ❐ Average active current of 1 mA at 3.4 MHz operation ❐ Average standby mode current of 250 µA ❐ Sleep mode current of 8 µA Industry standard configurations ❐ Operating voltages: • CY14C256I : VCC = 2.4 V to 2.6 V • CY14B256I : VCC = 2.7 V to 3.6 V • CY14E256I : VCC = 4.5 V to 5.5 V ❐ Industrial temperature ❐ 16-pin small outline integrated circuit (SOIC) package ❐ Restriction of hazardous substances (RoHS) compliant Overview The Cypress CY14C256I/CY14B256I/CY14E256I combines a 256-Kbit nvSRAM[1] with a full-featured RTC in a monolithic integrated circuit with serial I2C interface. The memory is organized as 64 K words of 8 bits each. The embedded nonvolatile elements incorporate the QuantumTrap technology, creating the world’s most reliable nonvolatile memory. The SRAM provides infinite read and write cycles, while the QuantumTrap cells provide highly reliable nonvolatile storage of data. Data transfers from SRAM to the nonvolatile elements (STORE operation) takes place automatically at power-down. On power-up, data is restored to the SRAM from the nonvolatile memory (RECALL operation). The STORE and RECALL operations can also be initiated by the user through I2C commands. Serial Number 8x8 VCC VCAP VRTCcap VRTCbat Manufacture ID/ Product ID Power Control Block Memory Control Register QuantrumTrap 32 K x 8 Command Register Sleep SDA SCL A2, A1, A0 WP Control Registers Slave 2 I C Control Logic Slave Address Decoder Memory Slave RTC Slave X in INT/SQW Xout Memory Address and Data Control SRAM 32 K x 8 STORE RECALL RTC Control Logic Registers Counters Note 1. Serial (I2C) nvSRAM will be referred to as nvSRAM throughout the datasheet. Cypress Semiconductor Corporation Document #: 001-65230 Rev. *B • 198 Champion Court • San Jose, CA 95134-1709 • 408-943-2600 Revised May 5, 2011 [+] Feedback CY14C256I CY14B256I, CY14E256I Contents Pinouts .............................................................................. 3 Pin Definitions .................................................................. 3 I2C Interface ...................................................................... 4 Protocol Overview ............................................................ 4 I2C Protocol – Data Transfer ....................................... 4 Data Validity ...................................................................... 5 START Condition (S) ........................................................ 5 STOP Condition (P) .......................................................... 5 Repeated START (Sr) ....................................................... 5 Byte Format ....................................................................... 5 Acknowledge / No-acknowledge ..................................... 5 High-Speed Mode (Hs-mode) .......................................... 6 Serial Data Format in Hs-mode ................................... 6 Slave Device Address ...................................................... 7 Memory Slave Device ................................................. 7 RTC Registers Slave Device ....................................... 7 Control Registers Slave Device ................................... 7 Memory Control Register ............................................ 8 Command Register ..................................................... 8 Write Protection (WP) ....................................................... 9 AutoStore Operation ........................................................ 9 Hardware STORE and HSB pin Operation ..................... 9 Hardware RECALL (Power Up) ................................... 9 Write Operation ............................................................... 10 Read Operation ............................................................... 10 Memory Slave Access .................................................... 10 Write nvSRAM ........................................................... 10 Current nvSRAM Read .............................................. 12 Random Address Read ............................................. 13 RTC Registers Slave Access ......................................... 14 Write RTC Registers ................................................. 14 Current Address RTC Registers Read ...................... 15 Random Address RTC Registers Read .................... 15 Control Registers Slave ................................................. 16 Write Control Registers ............................................. 16 Current Control Registers Read ................................ 17 Random Control Registers Read .............................. 17 Serial Number ................................................................. 18 Serial Number Write .................................................. 18 Serial Number Lock ................................................... 18 Serial Number Read .................................................. 18 Device ID Read ......................................................... 19 Executing Commands Using Command Register ..... 19 Document #: 001-65230 Rev. *B Real Time Clock Operation ............................................ 20 nvTIME Operation ..................................................... 20 Clock Operations ....................................................... 20 Reading the Clock ..................................................... 20 Setting the Clock ....................................................... 20 Backup Power ........................................................... 20 Stopping and Starting the Oscillator .......................... 20 Calibrating the Clock ................................................. 21 Alarm ......................................................................... 21 Watchdog Timer ........................................................ 21 Programmable Square Wave Generator ................... 22 Power Monitor ........................................................... 22 Backup Power Monitor .............................................. 22 Interrupts ................................................................... 22 Interrupt Register ....................................................... 22 Flags Register ........................................................... 23 Best Practices ................................................................. 29 Maximum Ratings ........................................................... 30 Operating Range ............................................................. 30 DC Electrical Characteristics ........................................ 30 Data Retention and Endurance ..................................... 31 Thermal Resistance ........................................................ 31 AC Test Conditions ........................................................ 32 RTC Characteristics ....................................................... 32 AC Switching Characteristics ....................................... 33 nvSRAM Specifications ................................................. 34 Software Controlled STORE/RECALL Cycles .............. 35 Hardware STORE Cycle ................................................. 36 Ordering Information ...................................................... 37 Ordering Code Definitions ......................................... 37 Package Diagram ............................................................ 38 Acronyms ........................................................................ 39 Document Conventions ................................................. 39 Units of Measure ....................................................... 39 Document History Page ................................................. 40 Sales, Solutions, and Legal Information ...................... 41 Worldwide Sales and Design Support ....................... 41 Products .................................................................... 41 PSoC Solutions ......................................................... 41 Page 2 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Pinouts Figure 1. Pin Diagram - 16-pin SOIC NC 1 16 VCC VRTCbat 2 15 INT/SQW Xout 3 14 VCAP Xin 4 13 A2 Top View not to scale WP 5 12 SDA A0 6 11 SCL VRTCcap 7 10 A1 8 9 HSB VSS Pin Definitions Pin Name I/O Type Description Clock: Runs at speeds up to a maximum of fSCL. SCL Input SDA Input/Output WP Input Write Protect: Protects the memory from all writes. This pin is internally pulled LOW and hence can be left open if not connected. A2-A0 Input Slave Address: Defines the slave address for I2C. These pins are internally pulled LOW and hence can be left open if not connected. HSB Input/Output Hardware STORE Busy: Output: Indicates busy status of nvSRAM when LOW. After each Hardware and Software STORE operation HSB is driven HIGH for a short time (tHHHD) with standard output high current and then a weak internal pull-up resistor keeps this pin HIGH (External pull-up resistor connection optional). Input: Hardware STORE implemented by pulling this pin LOW externally. VCAP Power supply AutoStore Capacitor: Supplies power to the nvSRAM during power loss to STORE data from the SRAM to nonvolatile elements. If not required, AutoStore must be disabled and this pin left as No Connect. It must never be connected to ground. VRTCcap Power supply Capacitor Backup for RTC: Left unconnected if VRTCbat is used. VRTCbat Power supply Battery Backup for RTC: Left unconnected if VRTCcap is used. Xout Output Xin Input INT/SQW Output NC No connect VSS Power supply Ground VCC Power supply Power supply I/O: Input/Output of data through I2C interface. Crystal output connection Crystal input connection Interrupt Output/Calibration/Square Wave. Programmable to respond to the clock alarm, the watchdog timer, and the power monitor. Also programmable to either active HIGH (push or pull) or LOW (open drain). In Calibration mode, a 512 Hz square wave is driven out. In Square Wave mode, the user may select a frequency of 1 Hz, 512 Hz, 4096 Hz, or 32768 Hz to be used as a continuous output. No connect. This pin is not connected to the die. Document #: 001-65230 Rev. *B Page 3 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I I2C Interface (0) operation. All signals are transmitted on the open-drain SDA line and are synchronized with the clock on SCL line. Each byte of data transmitted on the I2C bus is acknowledged by the receiver by holding the SDA line LOW on the ninth clock pulse. The request for write by the master is followed by the memory address and data bytes on the SDA line. The writes can be performed in burst-mode by sending multiple bytes of data. The memory address increments automatically after the receive/transmit of each byte on the falling edge of the ninth clock cycle. The new address is latched just prior to sending/receiving the acknowledgment bit. This allows the next sequential byte to be accessed with no additional addressing. On reaching the last memory location, the address rolls back to 0x0000 and writes continue. The slave responds to each byte sent by the master during a write operation with an ACK. A write sequence can be terminated by the master generating a STOP or Repeated START condition. I2C bus consists of two lines – serial clock line (SCL) and serial data line (SDA) – that carry information between multiple devices on the bus. I2C supports multi-master and multi-slave configurations. The data is transmitted from the transmitter to the receiver on the SDA line and is synchronized with the clock SCL generated by the master. The SCL and SDA lines are open-drain lines and are pulled up to VCC using resistors. The choice of a pull-up resistor on the system depends on the bus capacitance and the intended speed of operation. The master generates the clock, and all the data I/Os are transmitted in synchronization with this clock. The CY14X256I supports up to 3.4 MHz clock speed on SCL line. Protocol Overview A read request is performed at the current address location (address next to the last location accessed for read or write). The memory slave device responds to a read request by transmitting the data on the current address location to the master. A random address read may also be performed by first sending a write request with the intended address of read. The master must abort the write immediately after the last address byte and issue a Repeated START or STOP signal to prevent any write operation. The following read operation starts from this address. The master acknowledges the receipt of one byte of data by holding the SDA pin LOW for the ninth clock pulse. The reads can be terminated by the master sending a no-acknowledge (NACK) signal on the SDA line after the last data byte. The NACK signal causes the CY14X256I to release the SDA line and the master can then generate a STOP or a Repeated START condition to initiate a new operation. This device supports only a 7-bit addressable scheme. The master generates a START condition to initiate the communication followed by broadcasting a slave select byte. The slave select byte consists of a 7-bit slave address that the master intends to communicate with and R/W bit indicating a read or a write operation. The selected slave responds to this with an acknowledgement (ACK). After a slave is selected, the remaining part of the communication takes place between the master and the selected slave device. The other devices on the bus ignore the signals on the SDA line until a STOP or Repeated START condition is detected. The data transfer is done between the master and the selected slave device through the SDA pin synchronized with the SCL clock generated by the master. I2C Protocol – Data Transfer Each transaction in I2C protocol starts with the master generating a START condition on the bus, followed by a 7-bit slave address and eighth bit (R/W) indicating a read (1) or a write Figure 2. System Configuration using Serial (I2C) nvSRAM Vcc RPmin = (VCC - VOLmax) / IOL RPmax = tr / Cb SDA Microcontroller SCL Vcc Vcc A0 SCL A0 SCL A0 SCL A1 SDA A1 SDA A1 SDA WP A2 WP A2 CY14X256I #0 Document #: 001-65230 Rev. *B A2 WP CY14X256I CY14X256I #1 #7 Page 4 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Data Validity STOP Condition (P) The data on the SDA line must be stable during the HIGH period of the clock. The state of the data line can only change when the clock on the SCL line is LOW for the data to be valid. There are only two conditions under which the SDA line may change state with SCL line held HIGH: START and STOP condition. The START and STOP conditions are generated by the master to signal the beginning and end of a communication sequence on the I2C bus. A LOW to HIGH transition on the SDA line while SCL is HIGH indicates a STOP condition. This condition indicates the end of the ongoing transaction. START and STOP conditions are always generated by the master. The bus is considered to be busy after the START condition. The bus is considered to be free again after the STOP condition. Repeated START (Sr) START Condition (S) If a Repeated START condition is generated instead of a STOP condition, the bus continues to be busy. The ongoing transaction on the I2C lines is stopped and the bus waits for the master to send a slave ID for communication to restart. A HIGH to LOW transition on the SDA line while SCL is HIGH indicates a START condition. Every transaction in I2C begins with the master generating a START condition. Figure 3. START and STOP Conditions full pagewidth SDA SDA SCL SCL S P STOP Condition START Condition Figure 4. Data Transfer on the I2C Bus handbook, full pagewidth P SDA Acknowledgement signal from slave MSB SCL S or Sr 1 2 START or Repeated START condition 7 ACK Each operation in I2C is done using 8-bit words. The bits are sent in MSB first format on SDA line and each byte is followed by an ACK signal by the receiver. An operation continues till a NACK is sent by the receiver or STOP or Repeated START condition is generated by the master The SDA line must remain stable when the clock (SCL) is HIGH except for a START or STOP condition. Acknowledge / No-acknowledge After transmitting one byte of data or address, the transmitter releases the SDA line. The receiver pulls the SDA line LOW to acknowledge the receipt of the byte. Every byte of data transferred on the I2C bus needs a response with an ACK signal by the receiver to continue the operation. Failing to do so is considered as a NACK state. NACK is the state where receiver 2 3-8 9 ACK Byte complete, interrupt within slave Byte Format Document #: 001-65230 Rev. *B 1 9 8 Acknowledgement signal from receiver Clock line held LOW while interrupts are serviced Sr Sr or P STOP or Repeated START condition does not acknowledge the receipt of data and the operation is aborted. NACK can be generated by master during a READ operation in following cases: ■ The master did not receive valid data due to noise. ■ The master generates a NACK to abort the READ sequence. After a NACK is issued by the master, nvSRAM slave releases control of the SDA pin and the master is free to generate a Repeated START or STOP condition. NACK can be generated by nvSRAM slave during a WRITE operation in these cases: ■ nvSRAM did not receive valid data due to noise. ■ The master tries to access write protected locations on the nvSRAM. Master must restart the communication by generating a STOP or Repeated START condition. Page 5 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 5. Acknowledge on the I2C Bus handbook, full pagewidth DATA OUTPUT BY MASTER Not acknowledge (A) DATA OUTPUT BY SLAVE Acknowledge (A) SCL FROM MASTER 1 2 8 9 S Clock pulse for acknowledgement START Condition High-Speed Mode (Hs-mode) In Hs-mode, nvSRAM can transfer data at bit rates of up to 3.4 Mbit/s. A master code (0000 1XXXb) must be issued to place the device in high-speed mode. This enables master/slave communication for speeds up to 3.4 MHz. A stop condition will exit Hs-mode. Serial Data Format in Hs-mode Serial data transfer format in Hs-mode meets the standard-mode I2C-bus specification. Hs-mode can only commence after the following conditions (all of which are in F/S-modes): 1. START condition (S) 2. 8-bit master code (0000 1XXXb) 3. No-acknowledge bit (A) Single and multiple-byte reads and writes are supported. After the device enters into Hs-mode, data transfer continues in Hs-mode until stop condition is sent by master device. The slave switches back to F/S-mode after a STOP condition (P). To continue data transfer in Hs-mode, the master sends Repeated START (Sr). See Figure 13 on page 11 and Figure 16 on page 12 for Hs-mode timings for read and write operation. Figure 6. Data Transfer Format in Hs-mode handbook, full pagewidth F/S-mode S MASTER CODE Hs-mode A Sr SLAVE ADD. R/W A F/S-mode DATA n (bytes+ ack.) A/A P Hs-mode continues Sr SLAVE ADD. Document #: 001-65230 Rev. *B Page 6 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Slave Device Address Every slave device on an I2C bus has a device select address. The first byte after START condition contains the slave device address with which the master intends to communicate. The seven MSBs are the device address and the LSB (R/W bit) is used for indicating Read or Write operation. The CY14X256I reserves three sets of upper 4 MSBs [7:4] in the slave device address field for accessing the Memory, RTC Registers, and Control Registers. The accessing mechanism is described in the following section. The nvSRAM product provides three different functionalities: Memory, RTC Registers and Control Registers functions (such as serial number and product ID). The three functions of the device are accessed through different slave device addresses. The first four most significant bits [7:4] in the device address register are used to select between the nvSRAM functions. Table 1. Slave Device Addressing Bit 3 Bit 2 Bit 1 nvSRAM Bit 0 Function Select Bit 7 Bit 6 Bit 5 Bit 4 1 0 1 0 Device select ID R/W Selects Memory 1 1 0 1 Device select ID R/W CY14X256I Slave Devices Memory, 32 K × 8 Selects RTC Registers RTC Registers, 16 × 8 Control Registers - Memory Control Register, 1 × 8 0 0 1 1 Device select ID R/W - Serial Number, 8 × 8 Selects Control Registers - Device ID, 4 × 8 - Command Register, 1 × 8 Memory Slave Device RTC Registers Slave Device The nvSRAM device is selected for read/write if the master issues the slave address as 1010b followed by three bits of device select. If the slave address sent by the master matches with the Memory Slave device address then depending on the R/W bit of the slave address, the data will be either read from (R/W = ‘1’) or written to (R/W = ‘0’) the nvSRAM. The RTC Registers is selected for read/write if the master issues the slave address as 1101b followed by three bits of device select. Then, depending on the R/W bit of the slave address, data is either read from (R/W = ‘1’) or written to (R/W = ‘0’) the RTC Registers. The RTC Registers slave address is followed by one byte address of RTC Register for read/write operation. The RTC Registers map is explained in the Table 10. The address length for CY14X256I is 15 bits, and thus it requires two address bytes to map the entire memory address location. The dedicated two address bytes represent bit A0 to A14. However, since the address is only 15-bits, it implies that the first MSB bit that is fed in is ignored by the device. Although this bit is ‘don’t care’, Cypress recommends that this bit is treated as 0 to enable seamless transition to higher memory densities. Figure 7. Memory Slave Device Address MSB handbook, halfpage 1 1 Slave ID Document #: 001-65230 Rev. *B 0 A2 A1 A0 R/W Device Select MSB handbook, halfpage 1 LSB 1 0 1 Slave ID LSB 0 Figure 8. RTC Registers Slave Device Address A2 A1 A0 R/W Device Select Control Registers Slave Device The Control Registers Slave device includes the serial number, product ID, Memory Control, and Command Register. The nvSRAM Control Register Slave device is selected for read/write if the master issues the slave address as 0011b followed by three bits of device select. Then, depending on the R/W bit of the slave address, data is either read from (R/W = ‘1’) or written to (R/W = ‘0’) the device. Page 7 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 9. Control Registers Slave Device Address MSB handbook, halfpage 0 LSB 1 0 A2 1 A1 A0 R/W Device Select Slave ID Table 2. Control Registers Map Address 0x00 0x01 0x02 0x03 0x04 0x05 0x06 0x07 0x08 0x09 0x0A 0x0B 0x0C 0x0D 0xAA ■ Description Memory Control Register Serial Number 8 bytes Device ID Command Register Read/Write Details Read/Write Contains Block Protect bits and Serial Number lock bit Read/Write Programmable Serial (Read only Number. Locked by when SNL setting the Serial is set) Number lock bit in the Memory Control Register to ‘1’. The Command Register resides at address ‘AA’ of the Control Registers Slave device. This is a write only register. The byte written to this register initiates a STORE, RECALL, AutoStore Enable, AutoStore Disable, and Sleep mode operation as listed in Table 5. The section Executing Commands Using Command Register on page 19 explains how you can execute Command Register bytes. Table 5. Command Register Bytes Read only Device ID is factory programmed Reserved Command Register Reserved Reserved Write only Allows commands for STORE, RECALL, AutoStore Enable/Disable, SLEEP Mode The Memory Control Register contains the following bits: ■ Bit 5 0 Bit 4 0 Bit 3 BP1 (0) Bit 2 BP0 (0) Bit 1 0 Table 4. Block Protection Level 0 1/4 1/2 1 BP1:BP0 00 01 10 11 Block Protection None 0x6000-0x7FFF 0x4000-0x7FFF 0x0000-0x7FFF Document #: 001-65230 Rev. *B Description STORE 0110 0000 RECALL 0101 1001 0001 1001 1011 1001 ASENB ASDISB SLEEP STORE SRAM data to nonvolatile memory RECALL data from nonvolatile memory to SRAM Enable AutoStore Disable AutoStore Enter Sleep Mode for low power consumption ■ RECALL: Initiates nvSRAM Software RECALL. The nvSRAM cannot be accessed for tRECALL time after this instruction has been executed. The RECALL operation does not alter the data in the nonvolatile elements. A RECALL may be initiated in two ways: Hardware RECALL, initiated on power-up; and Software RECALL, initiated by a I2C RECALL instruction. ■ ASENB: Enables nvSRAM AutoStore. The nvSRAM cannot be accessed for tSS time after this instruction has been executed. This setting is not nonvolatile and needs to be followed by a manual STORE sequence if this is desired to survive the power cycle. The part comes from the factory with AutoStore Enabled. ■ ASDISB: Disables nvSRAM AutoStore. The nvSRAM cannot be accessed for tSS time after this instruction has been executed. This setting is not nonvolatile and needs to be followed by a manual STORE sequence if this is desired to survive power cycle. Bit 0 0 BP1:BP0: Block protect bits are used to protect 1/4, 1/2 or full memory array. These bits can be written through a write instruction to the 0x00 location of the Control Register Slave device. However, any STORE cycle transfers SRAM data into a nonvolatile cell regardless of whether or not the block is protected. The default value shipped from the factory for BP0 and BP1 is ‘0’. Command STORE: Initiates nvSRAM Software STORE. The nvSRAM cannot be accessed for tSTORE time after this instruction has been executed. When initiated, the device performs a STORE operation regardless of whether or not a write has been performed since the last NV operation. After the tSTORE cycle time is completed, the SRAM is activated again for read/write operations. Table 3. Memory Control Register Bits Bit 6 SNL (0) Data Byte [7:0] 0011 1100 ■ Memory Control Register Bit 7 0 S/N Lock (SNL) Bit: Serial Number Lock bit (SNL) is used to lock the serial number. After the bit is set to ‘1’, the serial number registers are locked and no modification is allowed. This bit cannot be cleared to ‘0’. The serial number is secured on the next STORE operation (Software STORE or AutoStore). If AutoStore is not enabled, user must perform the Software STORE operation to secure the lock bit status. If a STORE was not performed, the serial number lock bit will not survive the power cycle. The default value shipped from the factory for SNL is ‘0’. Note If AutoStore is disabled and VCAP is not required, it is required that the VCAP pin is left open. VCAP pin must never be connected to ground. Power Up RECALL operation cannot be disabled in any case. Page 8 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I ■ SLEEP: SLEEP instruction puts the nvSRAM in a sleep mode. When the SLEEP instruction is registered, the nvSRAM performs a STORE operation to secure the data to nonvolatile memory and then enters into sleep mode. Whenever nvSRAM enters into sleep mode, it initiates non volatile STORE cycle which results in losing an endurance cycle per sleep command execution. A STORE cycle starts only if a write to the SRAM has been performed since the last STORE or RECALL cycle. The nvSRAM enters into sleep mode in this manner: 1. The master sends a START command. 2. The master sends Control Registers Slave device ID with I2C write bit set (R/W = ’0’). 3. The slave (nvSRAM) sends an ACK back to the master. 4. The master sends Command Register address (0xAA). 5. The slave (nvSRAM) sends an ACK back to the master. 6. The master sends Command Register byte for entering into sleep mode. 7. The slave (nvSRAM) sends an ACK back to the master. 8. The master generates a STOP condition. Once in sleep mode, the device starts consuming IZZ current tSLEEP time after SLEEP instruction is registered. The device is not accessible for normal operations until it is out of sleep mode. The nvSRAM wakes up after tWAKE duration after the device slave address is transmitted by the master. Transmitting any of the three slave addresses wakes the nvSRAM from sleep mode. The nvSRAM device is not accessible during tSLEEP and tWAKE interval and any attempt to access the nvSRAM device by the master is ignored and nvSRAM sends NACK to the master. An alternate method of determining when the device is ready is for the master to send read or write commands and look for an ACK. Write Protection (WP) The Write Protect (WP) pin is an active HIGH pin and protects the entire memory and all registers from write operations. To inhibit all the write operations, this pin must be held HIGH. When this pin is HIGH, all memory and register writes are prohibited and the address counter is not incremented. This pin is internally pulled LOW and, therefore, can be left open if not used. AutoStore Operation The AutoStore operation is a unique feature of nvSRAM that automatically stores the SRAM data to QuantumTrap cells during power-down. This STORE makes use of an external capacitor (VCAP) and enables the device to safely STORE the data in the nonvolatile memory when power goes down. During normal operation, the device draws current from VCC to charge the capacitor connected to the VCAP pin. When the voltage on the VCC pin drops below VSWITCH during power-down, the device inhibits all memory accesses to nvSRAM and automatically performs a conditional STORE operation using the charge from the VCAP capacitor. The AutoStore operation is not initiated if no write cycle has been performed since the last STORE or RECALL. Note If a capacitor is not connected to VCAP pin, AutoStore must be disabled by issuing the AutoStore Disable instruction Document #: 001-65230 Rev. *B specified in Command Register on page 8. If AutoStore is enabled without a capacitor on VCAP pin, the device attempts an AutoStore operation without sufficient charge to complete the Store. This will corrupt the data stored in nvSRAM as well as the serial number and it will unlock the SNL bit. Figure 10 shows the proper connection of the storage capacitor (VCAP) for AutoStore operation. See the DC Electrical Characteristics on page 30 for the size of the VCAP. Figure 10. AutoStore Mode VCC 0.1 uF VCC VCAP VSS VCAP Hardware STORE and HSB pin Operation The HSB pin in CY14X256I is used to control and acknowledge STORE operations. If no STORE or RECALL is in progress, this pin can be used to request a Hardware STORE cycle. When the HSB pin is driven LOW, the device conditionally initiates a STORE operation after tDELAY duration. An actual STORE cycle starts only if a write to the SRAM has been performed since the last STORE or RECALL cycle. Reads and Writes to the memory are inhibited for tSTORE duration or as long as HSB pin is LOW. The HSB pin also acts as an open drain driver (internal 100 k weak pull-up resistor) that is internally driven LOW to indicate a busy condition when the STORE (initiated by any means) is in progress. Note After each Hardware and Software STORE operation HSB is driven HIGH for a short time (tHHHD) with standard output high current and then remains HIGH by internal 100 k pull-up resistor. Note For successful last data byte STORE, a hardware STORE should be initiated at least one clock cycle after the last data bit D0 is received. Upon completion of the STORE operation, the nvSRAM memory access is inhibited for tLZHSB time after HSB pin returns HIGH. Leave the HSB pin unconnected if not used. Hardware RECALL (Power Up) During power-up, when VCC crosses VSWITCH, an automatic RECALL sequence is initiated that transfers the content of nonvolatile memory to the SRAM. The data may have been previously stored on the nonvolatile memory through a STORE sequence. Page 9 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I A Power Up RECALL cycle takes tFA time to complete and the memory access is disabled during this time. HSB pin can be used to detect the ready status of the device. Write Operation The last bit of the slave device address indicates a read or a write operation. In case of a write operation, the slave device address is followed by the memory or register address and data. A write operation continues as long as a STOP or Repeated START condition is generated by the master or if a NACK is issued by the nvSRAM. A NACK is issued from the nvSRAM under the following conditions: 1. A valid Device ID is not received. 2. A write (burst write) access to a protected memory block address returns a NACK from nvSRAM after the data byte is received. However, the address counter is set to this address and the following current read operation starts from this address. 3. A write/random read access to an invalid or out-of-bound memory address returns a NACK from the nvSRAM after the address is received. The address counter remains unchanged in such a case. 4. A write to the Command Register with an invalid command. This operation returns a NACK from the nvSRAM. After a NACK is sent out from the nvSRAM, the write operation is terminated and any data on the SDA line is ignored till a STOP or a Repeated START condition is generated by the master. For example, consider a case where the burst write access is performed on Control Register Slave address 0x01 for writing the serial number and continued to the address 0x09, which is a read-only register. The device returns a NACK and address counter is not incremented. A following read operation is started from the address 0x09. Further, any write operation which starts from a write protected address (say, 0x09) is responded by the nvSRAM with a NACK after the data byte is sent and set the address counter to this address. A following read operation starts from the address 0x09 in this case also. Note In case you try to read/write access an address that does not exist (for example 0x0D in Control Register Slave or 0x3F in RTC registers), nvSRAM responds with a NACK immediately after the out-of-bound address is transmitted. The address counter remains unchanged and holds the previous successful read or write operation address. A write operation is performed internally with no delay after the eighth bit of data is transmitted. If a write operation is not intended, the master must terminate the write operation before the eighth clock cycle by generating a STOP or Repeated START condition. More details on write instructions are provided in the section Memory Slave Access. Document #: 001-65230 Rev. *B Read Operation If the last bit of the slave device address is ‘1’, a read operation is assumed and the nvSRAM takes control of the SDA line immediately after the slave device address byte is sent out by the master. The read operation starts from the current address location (the location following the previous successful write or read operation). When the last address is reached, the address counter loops back to the first address. In case of the Control Register Slave, whenever a burst read is performed such that it flows to a non-existent address, the reads operation loops back to 0x00. This is applicable, in particular, for the Command Register. Read operation can be ended using the following methods: 1. The master issues a NACK on the ninth clock cycle followed by a STOP or a Repeated START condition on the tenth clock cycle. 2. The master generates a STOP or Repeated START condition on the ninth clock cycle. More details on write instruction are provided in the section Memory Slave Access. Memory Slave Access The following sections describe the data transfer sequence required to perform read or write operations from nvSRAM. Write nvSRAM Each write operation consists of a slave address being transmitted after the start condition. The last bit of slave address must be set as ‘0’ to indicate a Write operation. The master may write one byte of data or continue writing multiple consecutive address locations while the internal address counter keeps incrementing automatically. The address register is reset to 0x0000 after the last address in memory is accessed. The write operation continues till a STOP or Repeated START condition is generated by the master or a NACK is issued by the nvSRAM. A write operation is executed only after nvSRAM receives all the eight data bits. The nvSRAM sends an ACK signal after a successful write operation. A write operation may be terminated by the master by generating a STOP condition or a Repeated START operation. If the master desires to abort the current write operation without altering the memory contents, this should be done using a START/STOP condition prior to the eighth data bit. If the master tries to access a write protected memory address on the nvSRAM, a NACK is returned after the data byte intended to write the protected address is transmitted and address counter will not be incremented. Similarly, in a burst mode write operation, a NACK is returned when the data byte that attempts to write a protected memory location and the address counter is not incremented. Page 10 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 11. Single-Byte Write into nvSRAM (except Hs-mode) S T A R T By Master SDA Line Address MSB Memory Slave Address S 1 0 1 0 A2 A1 A0 0 Address LSB S T 0 P Data Byte P X By nvSRAM A A A A Figure 12. Multi-Byte Write into nvSRAM (except Hs-mode) S T A R Memory Slave Address T SDA Line S 1 0 1 Address MSB 0 A2 A1 A0 0 Address LSB Data Byte N Data Byte 1 ~ ~ By Master S T 0 P X P By nvSRAM A A A A A Figure 13. Single-Byte Write into nvSRAM (Hs-mode) By Master SDA Line S T A R T Hs-mode command S 0 0 0 0 1 Memory Slave Address X X X Sr 1 0 Address MSB 1 0 A2 A1 A0 0 S T 0 P Data Byte Address LSB P X By nvSRAM A A A A A Figure 14. Multi-Byte Write into nvSRAM (Hs-mode) SDA Line Hs-mode command S 0 0 0 0 1 Memory Slave Address X X X Sr 1 0 Address MSB 1 0 A2 A1 A0 0 Data Byte 1 Address LSB ~ ~ By Master S T A R T X By nvSRAM A By Master A By nvSRAM Document #: 001-65230 Rev. *B ~ ~ SDA Line A A A S T 0 P Data Byte N Data Byte 3 Data Byte 2 A A P A Page 11 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Current nvSRAM Read terminate a read operation after reading 1 byte or continue reading addresses sequentially till the last address in the memory after which the address counter rolls back to the address 0x0000. The valid methods of terminating read access are described in Section Read Operation on page 10. Each read operation starts with the master transmitting the nvSRAM slave address with the LSB set to ‘1’ to indicate ‘Read’. The reads start from the address on the address counter. The address counter is set to the address location next to the last accessed with a ‘Write’ or ‘Read’ operation. The master may Figure 15. Current Location Single-Byte nvSRAM Read (except Hs–mode) S T A R T By Master SDA Line A Memory Slave Address S 1 0 1 0 S T 0 P P A2 A1 A0 1 By nvSRAM Data Byte A Figure 16. Current Location Multi-Byte nvSRAM Read (except Hs–mode) SDA Line S S T 0 P A A Memory Slave Address 1 0 1 0 A2 A1 A0 1 P ~ ~ By Master S T A R T By nvSRAM Data Byte N Data Byte A Figure 17. Current Location Single-Byte nvSRAM Read (Hs–mode) S T A R T By Master SDA Line Hs-mode command S 0 0 0 0 1 S A T 0 P Memory Slave Address X X X Sr 1 0 P 1 0 A2 A1 A0 1 By nvSRAM Data Byte A A Figure 18. Current Location Multi-Byte nvSRAM Read (Hs–mode) SDA Line A Hs-mode command S 0 0 0 0 1 X X X Sr 1 0 1 0 A2 A1 A0 1 Data Byte By nvSRAM A Document #: 001-65230 Rev. *B A Memory Slave Address ~ ~ By Master S T A R T S T 0 P P Data Byte N A Page 12 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Random Address Read A random address read is performed by first initiating a write operation and generating a Repeated START immediately after the last address byte is acknowledged. The address counter is set to this address and the next read access to this slave initiates read operation from here. The master may terminate a read operation after reading 1 byte or continue reading addresses sequentially till the last address in the memory after which the address counter rolls back to the start address 0x0000. Figure 19. Random Address Single-Byte Read (except Hs–mode) S T A R T By Master SDA Line A Memory Slave Address S 1 1 0 Address MSB 0 A2 A1 A0 0 Address LSB Memory slave Address 0 Sr 1 X 1 S T 0 P P 0 A2 A1 A0 1 By nvSRAM Data Byte A A A A Figure 20. Random Address Multi-Byte Read (except Hs–mode) By Master SDA Line S A Memory Slave Address 1 0 1 Address MSB 0 A2 A1 A0 0 Address LSB Memory slave Address Sr 1 X 0 1 ~ ~ S T A R T 0 A2 A1 A0 1 By nvSRAM Data Byte 1 A A A S T 0 P A A P Data Byte N Figure 21. Random Address Single-Byte Read (Hs–mode) SDA Line Hs-mode command S 0 0 0 0 1 Address MSB Memory Slave Address X X X Sr 1 0 1 0 A2 A1 A0 0 Address LSB Memory Slave Address Sr 1 0 X 1 0 A2 A1 A0 1 ~ ~ By Master S T A R T By nvSRAM A A A A A S T A 0 P P Data Byte Document #: 001-65230 Rev. *B Page 13 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 22. Random Address Multi-Byte Read (Hs–mode) SDA Line Hs-mode command S 0 0 0 0 1 Address MSB Memory Slave Address X X X Sr 1 0 1 0 A2 A1 A0 0 Address LSB Memory Slave Address Sr 1 0 X 1 0 A2 A1 A0 1 ~ ~ By Master S T A R T By nvSRAM A A A A S T 0 P P ~ ~ Data Byte A A A Data Byte N RTC Registers Slave Access generated by the master or a NACK is issued by the nvSRAM RTC Registers Slave. The following sections describe the data transfer sequence required to perform read or write operations from RTC registers. A write operation is executed only after all the eight data bits have been received by the nvSRAM. The nvSRAM sends an ACK signal after the successful operation of the write instruction A write operation may be terminated by the master by generating a STOP condition or a Repeated START operation before the last data bit is sent. Write RTC Registers A write to RTC registers is initiated with the RTC Registers Slave address followed by one byte of address and data. The master may write one byte of data or continue writing multiple consecutive address locations while the internal address counter keeps incrementing automatically. The address register is reset to 0x00 after the last RTC register is accessed. The write operation continues till a STOP or Repeated START condition is If the master tries to access an out of bound memory address on the RTC Registers Slave, a NACK is returned after the address byte is transmitted. The address counter remains unaffected and the following current read operation starts from the address value held in the address counter. Figure 23. Single-Byte Write into RTC Registers S T A R T By Master SDA Line S RTC Registers Slave Address 1 1 0 RTC Register Address S T 0 P Data Byte P 1 A2 A1 A0 0 By nvSRAM A A A Figure 24. Multi-Byte Write into RTC Registers SDA Line S RTC Register Address RTC Registers Slave Address 1 1 0 1 A2 A1 A0 0 By nvSRAM A Document #: 001-65230 Rev. *B S T 0 P Data Byte N Data Byte P ~ ~ By Master S T A R T A A A Page 14 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Current Address RTC Registers Read first location (0x00) and read operation continues. The master may terminate a read operation after reading one byte or continue reading addresses sequentially till the last address in the memory after which the address counter rolls back to the address 0x00. A read operation may be terminated by the master by generating a STOP condition or a Repeated START operation or a NACK. A current read of RTC registers starts with the master sending the RTC Registers Slave address after the START condition. All read operations begin from the current address (the address next to previously accessed address location). After the last address is read sequentially, the address latch loops back to the Figure 25. Current Address RTC Registers Single-Byte Read S T A R T By Master SDA Line S A RTC Registers Slave Address 1 1 0 S T 0 P P 1 A2 A1 A0 1 By nvSRAM Data Byte A Figure 26. Current Address RTC Registers Multi-Byte Read By Master SDA Line S A RTC Registers Slave Address 1 1 0 1 A2 A1 A0 1 A By nvSRAM Data Byte 1 Random Address RTC Registers Read A S T 0 P P ~ ~ S T A R T Data Byte N address counter rolls back to the start address location of RTC (0x00). A random address read is performed by first initiating a write operation and generating a Repeated START immediately after the last address byte is acknowledged. The address counter is set to this address and the next read access to this slave initiates the read operation from here. The master may terminate a read operation after reading one byte or continue reading addresses sequentially till the last address in the memory after which the A random address read attempt on an out of bound memory address on the RTC Registers Slave is responded back with a NACK from the nvSRAM after the address byte is transmitted. The address counter remains unaffected and the following current read operation starts from the address value held in the address counter. Figure 27. Random Address RTC Registers Single-Byte Read S T A R T By Master SDA Line S RTC Register Address RTC Registers Slave Address 1 1 0 1 RTC Registers Slave Address A2 A1 A0 0 Sr 1 1 0 1 A2 A1 A0 S T 0 P A P 1 By nvSRAM Data Byte A A A Figure 28. Random Address RTC Registers Multi-Byte Read RTC Registers Slave Address SDA Line S 1 1 0 A RTC Register Address 1 A2 A1 A0 0 Sr 1 1 0 1 A2 A1 A0 1 BynvSRAM Data Byte 1 A Document #: 001-65230 Rev. *B A RTC Registers Slave Address A ~ ~ By Master S T A R T S T 0 P P Data Byte N A Page 15 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Control Registers Slave The following sections describe the data transfer sequence required to perform read or write operations from Control Registers Slave. Write Control Registers To write the Control Registers Slave, the master transmits the Control Registers Slave address after generating the START condition. The write sequence continues from the address location specified by the master till the master generates a STOP condition or the last writable address location. If a non-writable address location is accessed for write operation during a normal write or a burst, the slave generates a NACK after the data byte is sent and the write sequence terminates. Any following data bytes are ignored and the address counter is not incremented. If a write operation is performed on the Command Register (0xAA), the following current read operation also begins from the first address (0x00) as in this case, the current address is an out-of-bound address. The address is not incremented and the next current read operation begins from this address location. If a write operation is attempted on an out-of-bound address location, the nvSRAM sends a NACK immediately after the address byte is sent. Further, if the serial number is locked, only two addresses (0xAA or Command Register, and 0x00 or Memory Control Register) are writable in the Control Registers Slave. On a write operation to any other address location, the device will acknowledge command byte and address bytes but it returns a NACK from the control Registers Slave for data bytes. In this case, the address will not be incremented and a current read will happen from the last acknowledged address. The nvSRAM Control Registers Slave sends a NACK when an out of bound memory address is accessed for write operation, by the master. In such a case, a following current read operation begins from the last acknowledged address. Figure 29. Single-Byte Write into Control Registers S T A R T By Master SDA Line S Control Registers Slave Address 0 0 1 Control Register Address S T 0 P Data Byte P 1 A2 A1 A0 0 By nvSRAM A A A Figure 30. Multi-Byte Write into Control Registers SDA Line S Control Registers Slave Address 0 0 1 Control Register Address Data Byte 1 A2 A1 A0 0 By nvSRAM A Document #: 001-65230 Rev. *B S T 0 P Data Byte N P ~ ~ By Master S T A R T A A A Page 16 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Current Control Registers Read address location and loops back to the first location (0x00). Note that the Command Register is a write only register and is not accessible through the sequential read operations. If a burst read operation begins from the Command Register (0xAA), the address counter wraps around to the first address in the register map (0x00). A read of Control Registers Slave is started with master sending the Control Registers Slave address after the START condition with the LSB set to ‘1’. The reads begin from the current address which is the next address to the last accessed location. The reads to Control Registers Slave continues until the last readable Figure 31. Control Registers Single-Byte Read S T A R T By Master SDA Line S Control Registers Slave Address 0 0 1 A S T 0 P P 1 A2 A1 A0 1 By nvSRAM Data Byte A Figure 32. Current Control Registers Multi-Byte Read S T A R T SDA Line S Control Registers Slave Address 0 0 1 A A 1 A2 A1 A0 1 By nvSRAM Data Byte S T 0 P P ~ ~ By Master Data Byte N A Random Control Registers Read A read of random address may be performed by initiating a write operation to the intended location of read and immediately following with a Repeated START operation. The reads to Control Registers Slave continues till the last readable address location and loops back to the first location (0x00). Note that the Command Register is a write only register and is not accessible through the sequential read operations. A random read starting at the Command Register (0xAA) loops back to the first address in the Control Register register map (0x00). If a random read operation is initiated from an out-of-bound memory address, the nvSRAM sends a NACK after the address byte is sent. Figure 33. Random Control Registers Single-Byte Read By Master SDA Line S T A R T S Control Registers Slave Address 0 0 1 Control Register Address A Control Registers Slave Address Sr 0 1 A2 A1 A0 0 0 1 1 A2 A1 A0 1 S T 0 P P By nvSRAM Data Byte A Document #: 001-65230 Rev. *B A A Page 17 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 34. Random Control Registers Multi-Byte Read SDA Line S Control Registers Slave Address 0 0 1 Control Register Address A Control Registers Slave Address Sr 0 1 A2 A1 A0 0 0 1 1 A2 A1 A0 1 ~ ~ By Master S T A R T By nvSRAM Data Byte A A A A S T 0 P P Data Byte N Serial Number Serial number is an 8 byte memory space provided to the user to uniquely identify this device. It typically consists of a two byte customer ID, followed by five bytes of unique serial number and one byte of CRC check. However, nvSRAM does not calculate the CRC and it is up to the user to utilize the eight byte memory space in the desired format. The default values for the eight byte locations are set to ‘0x00’. Serial Number Write The serial number can be accessed through the Control Registers Slave Device. To write the serial number, master transmits the Control Registers Slave address after the START condition and writes to the address location from 0x01 to 0x08. The content of Serial Number registers is secured to nonvolatile memory on the next STORE operation. If AutoStore is enabled, nvSRAM automatically stores the serial number in the nonvolatile memory on power-down. However, if AutoStore is disabled, user must perform a STORE operation to secure the contents of Serial Number registers. Note If the serial number lock (SNL) bit is not set, the serial number registers can be re-written regardless of whether or not a STORE has happened. After the serial number lock bit is set, no writes to the serial number registers are allowed. If the master tries to perform a write operation to the serial number registers Document #: 001-65230 Rev. *B when the lock bit is set, a NACK is returned and write is not performed. Serial Number Lock After writes to serial number registers is complete, the master is responsible for locking the serial number by setting the serial number lock bit to ‘1’ in the Memory Control Register (0x00). The content of Memory Control Register and serial number are secured on the next STORE operation (STORE or AutoStore). If AutoStore is not enabled, user must perform the STORE operation to secure the lock bit status. If a STORE was not performed, the serial number lock bit will not survive the power cycle. The serial number lock bit and 8-byte serial number is defaults to ‘0’ at power-up. Serial Number Read Serial number can be read back by a read operation of the intended address of the Control Registers Slave. The Control Registers Device loops back from the last address (excluding the Command Register) to 0x00 address location while performing burst read operation. The serial number resides in the locations from 0x01 to 0x08. Even if the serial number is not locked, a serial number read operation returns the current values written to the serial number registers. The master may perform a serial number read operation to confirm if the correct serial number is written to the registers before setting the lock bit. Page 18 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Device ID Read registers are set in the factory and are read only registers for the user. Device ID is a 4-byte code consisting of JEDEC assigned manufacturer ID, product ID, density ID, and die revision. These Table 6. Device ID Bits #of Bits 31 - 21 (11 bits) 20 - 7 (14 bits) 6-3 (4 bits) 2-0 (3 bits) Device Manufacturer ID Product ID Density ID Die Rev CY14C256I 00000110100 00001111000001 0010 000 CY14B256I 00000110100 00001111010001 0010 000 CY14E256I 00000110100 00001111100101 0010 000 The device ID is divided into four parts as shown in Table 6: 4. Die Rev (3 bits) 1. Manufacturer ID (11 bits) This is used to represent any major change in the design of the product. The initial setting of this is always 0x0. This is the JEDEC assigned manufacturer ID for Cypress. JEDEC assigns the manufacturer ID in different banks. The first three bits of the manufacturer ID represent the bank in which ID is assigned. The next eight bits represent the manufacturer ID. Executing Commands Using Command Register The Control Registers Slave allows different commands to be executed by writing the specific command byte in the Command Register (0xAA). The command byte codes for each command are specified in Table 5. During the execution of these commands the device is not accessible and returns a NACK if any of the three slave devices is selected. If an invalid command is sent by the master, nvSRAM responds with a NACK indicating that the command was not successful. The address latch of this slave continues to point to the Command Register address. Cypress manufacturer ID is 0x34 in bank 0. Therefore the manufacturer ID for all Cypress nvSRAM products is as given below: Cypress ID - 000_0011_0100 2. Product ID (14 bits) The product ID for device is shown in the Table 6. 3. Density ID (4 bits) The 4-bit density ID is used as shown in Table 6 for indicating the 256 Kb density of the product. Figure 35. Command Execution using Command Register By Master SDA Line S T A R T S Control Register Slave Address 0 0 1 Command Register Address 1 A2 A1 A0 0 1 0 1 0 1 0 1 S T O P Command Byte P 0 By nvSRAM A Document #: 001-65230 Rev. *B A A Page 19 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Real Time Clock Operation nvTIME Operation The CY14X256I offers internal registers that contain clock, alarm, watchdog, interrupt, and control functions. The RTC registers occupy a separate address space from nvSRAM and are accessible through the Read RTC register and Write RTC register sequence on register addresses 0x00 to 0x0F. Internal double buffering of the clock and the timer information registers prevents accessing transitional internal clock data during a read or write operation. Double buffering also circumvents disrupting normal timing counts or the clock accuracy of the internal clock when accessing clock data. Clock and alarm registers store data in BCD format. Clock Operations The clock registers maintain time up to 9,999 years in one-second increments. The time can be set to any calendar time and the clock automatically keeps track of days of the week and month, leap years, and century transitions. There are eight registers dedicated to the clock functions, which are used to set time with a write cycle and to read time during a read cycle. These registers contain the time of day in BCD format. Bits defined as ‘0’ are currently not used and are reserved for future use by Cypress. Reading the Clock The double buffered RTC register structure reduces the chance of reading incorrect data from the clock. Stop internal updates to the CY14X256I time keeping registers before reading clock data to prevent reading of data in transition. Stopping the register updates does not affect clock accuracy. When an read sequence of RTC device is initiated, the update of the user timekeeping registers stops and does not restart until a STOP or a Repeated START condition is generated. The RTC registers are read while the internal clock continues to run. After the end of read sequence, all the RTC registers are simultaneously updated within 20 ms. Setting the Clock A write access to the RTC device stops updates to the time keeping registers and enables the time to be set. The correct day, date, and time is then written into the registers and must be in 24-hour BCD format. The time written is referred to as the “Base Time”. This value is stored in nonvolatile registers and used in the calculation of the current time. When a STOP or a Repeated START condition is encountered, the values of timekeeping registers are transferred to the actual clock counters after which the clock resumes normal operation. If a valid STOP or Repeated START condition is not generated by the master, the time written to the RTC registers is never transferred to the actual clock counters. If the time written to the timekeeping registers is not in the correct BCD format, each invalid nibble of the RTC registers continue counting to 0xF before rolling over to 0x0 after which RTC resumes normal operation. Note After ‘W’ bit is set to ‘0’, values written into the timekeeping, alarm, calibration, and interrupt registers are transferred to the RTC time keeping counters in tRTCp time. These counter values must be saved to nonvolatile memory either by initiating a Document #: 001-65230 Rev. *B Software/Hardware STORE or AutoStore operation. While working in AutoStore disabled mode, perform a STORE operation after tRTCp time while writing into the RTC registers for the modifications to be correctly recorded. Backup Power The RTC in the CY14X256I is intended for permanently powered operation. The VRTCcap or VRTCbat pin is connected depending on whether a capacitor or battery is chosen for the application. When the primary power, VCC, fails and drops below VSWITCH the device switches to the backup power supply. The clock oscillator uses very little current, which maximizes the backup time available from the backup source. Regardless of the clock operation with the primary source removed, the data stored in the nvSRAM is secure, having been stored in the nonvolatile elements when power was lost. During backup operation, the CY14X256I consumes a 0.45 µA (Typ) at room temperature. The user must choose capacitor or battery values according to the application. Backup time values based on maximum current specifications are shown in the following table. Nominal backup times are approximately two times longer. Table 7. RTC Backup Time Capacitor Value Backup Time (CY14B256I) 0.1F 60 hours 0.47F 12 days 1.0F 25 days Using a capacitor has the obvious advantage of recharging the backup source each time the system is powered up. If a battery backup is used, a 3-V lithium battery is recommended and the CY14X256I sources current only from the battery when the primary power is removed. However, the battery is not recharged at any time by the CY14X256I. The battery capacity must be chosen for total anticipated cumulative down time required over the life of the system. Stopping and Starting the Oscillator The OSCEN bit in the calibration register at 0x08 controls the enable and disable of the oscillator. This bit is nonvolatile and is shipped to customers in the “enabled” (set to ‘0’) state. To preserve the battery life when the system is in storage, OSCEN must be set to ‘1’. This turns off the oscillator circuit, extending the battery life. If the OSCEN bit goes from disabled to enabled, it takes approximately one second (two seconds maximum) for the oscillator to start. While system power is off, if the voltage on the backup supply (VRTCcap or VRTCbat) falls below their respective minimum level, the oscillator may fail.The CY14X256I has the ability to detect oscillator failure when system power is restored. This is recorded in the Oscillator Fail Flag (OSCF) of the flags register at the address 0x00. When the device is powered on (VCC goes above VSWITCH) the OSCEN bit is checked for the ‘enabled’ status. If the OSCEN bit is enabled and the oscillator is not active within the first 5 ms, the OSCF bit is set to ‘1’. The system must check for this condition and then write ‘0’ to clear the flag. Page 20 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Note that in addition to setting the OSCF flag bit, the time registers are reset to the ‘Base Time’ (when a read sequence of the RTC device is initiated, the update of the user timekeeping registers stops and does not restart until a STOP or a Repeated START condition is generated. The RTC registers are read while the internal clock continues to run. After the end of read sequence, all the RTC registers are simultaneously updated within 20 ms.), which is the value last written to the timekeeping registers. The control or calibration registers and the OSCEN bit are not affected by the ‘oscillator failed’ condition. The value of OSCF must be reset to ‘0’ when the time registers are written for the first time. This initializes the state of this bit which may have become set when the system was first powered on. To reset OSCF, set the write bit ‘W’ (in the flags register at 0x00) to a ‘1’ to enable writes to the flags register. Write a ‘0’ to the OSCF bit and then reset the write bit to ‘0’ to disable writes. Calibrating the Clock The RTC is driven by a quartz-controlled crystal with a nominal frequency of 32.768 kHz. Clock accuracy depends on the quality of the crystal and calibration. The crystals available in the market typically have an error of +20 ppm to +35 ppm. However, CY14X256I employs a calibration circuit that improves the accuracy to +1/–2 ppm at 25 °C. This implies an error of +2.5 seconds to -5 seconds per month. The calibration circuit adds or subtracts counts from the oscillator divider circuit to achieve this accuracy. The number of pulses that are suppressed (subtracted, negative calibration) or split (added, positive calibration) depends upon the value loaded into the five calibration bits found in the calibration register at 0x08. The calibration bits occupy the five lower order bits in the calibration register. These bits are set to represent any value between ‘0’ and 31 in binary form. Bit D5 is a sign bit, where a ‘1’ indicates positive calibration and a ‘0’ indicates negative calibration. Adding counts speeds the clock up and subtracting counts slows the clock down. If a binary ‘1’ is loaded into the register, it corresponds to an adjustment of 4.068 or –2.034 ppm offset in oscillator error, depending on the sign. Calibration occurs within a 64-minute cycle. The first 62 minutes in the cycle may, once per minute, have one second shortened by 128 or lengthened by 256 oscillator cycles. If a binary ‘1’ is loaded into the register, only the first two minutes of the 64-minute cycle are modified. If a binary 6 is loaded, the first 12 are affected, and so on. Therefore, each calibration step has the effect of adding 512 or subtracting 256 oscillator cycles for every 125,829,120 actual oscillator cycles, that is, 4.068 or –2.034 ppm of adjustment per calibration step in the calibration register. To determine the required calibration, the CAL bit in the flags register (0x00) must be set to ‘1’. This causes the INT pin to toggle at a nominal frequency of 512 Hz. Any deviation measured from the 512 Hz indicates the degree and direction of the required correction. For example, a reading of 512.01024 Hz indicates a +20 ppm error. Hence, a decimal value of –10 (001010b) must be loaded into the Calibration register to offset this error. Note Setting or changing the calibration register does not affect the test output frequency. Document #: 001-65230 Rev. *B To set or clear CAL, set the write bit ‘W’ (in the flags register at 0x00) to ‘1’ to enable writes to the flags register. Write a value to CAL, and then reset the write bit to ‘0’ to disable writes. Alarm The alarm function compares user programmed values of alarm time and date (stored in the registers 0x01-5) with the corresponding time of day and date values. When a match occurs, the alarm internal flag (AF) is set and an interrupt is generated on INT pin if alarm interrupt enable (AIE) bit is set. There are four alarm match fields - date, hours, minutes, and seconds. Each of these fields has a match bit that is used to determine if the field is used in the alarm match logic. Setting the match bit to ‘0’ indicates that the corresponding field is used in the match process. Depending on the match bits, the alarm occurs as specifically as once a month or as frequently as once every minute. Selecting none of the match bits (all 1s) indicates that no match is required and therefore, alarm is disabled. Selecting all match bits (all 0s) causes an exact time and date match. There are two ways to detect an alarm event: by reading the AF flag or monitoring the INT pin. The AF flag in the flags register at 0x00 indicates that a date or time match has occurred. The AF bit is set to ‘1’ when a match occurs. Reading the flags register clears the alarm flag bit (and all others). A hardware interrupt pin may also be used to detect an alarm event. To set, clear or enable an alarm, set the ‘W’ bit (in flags register - 0x00) to ‘1’ to enable writes to alarm registers. After writing the alarm value, clear the ‘W’ bit back to ‘0’ for the changes to take effect. Note CY14X256I requires the alarm match bit for seconds (0x02 - D7) to be set to ‘0’ for proper operation of Alarm Flag and Interrupt. Watchdog Timer The watchdog timer is a free running down counter that uses the 32 Hz clock (31.25 ms) derived from the crystal oscillator. The oscillator must be running for the watchdog to function. It begins counting down from the value loaded in the watchdog timer register. The timer consists of a loadable register and a free running counter. On power-up, the watchdog time out value in register 0x07 is loaded into the Counter Load register. Counting begins on power-up and restarts from the loadable value any time the Watchdog Strobe (WDS) bit is set to ‘1’. The counter is compared to the terminal value of ‘0’. If the counter reaches this value, it causes an internal flag and an optional interrupt output. You can prevent the time out interrupt by setting WDS bit to ‘1’ prior to the counter reaching ‘0’. This causes the counter to reload with the watchdog time out value and to be restarted. As long as the user sets the WDS bit prior to the counter reaching the terminal value, the interrupt and WDT flag never occur. New time out values are written by setting the watchdog write bit to ‘0’. When the WDW is ‘0’, new writes to the watchdog time out value bits D5-D0 are enabled to modify the time out value. When WDW is ‘1’, writes to bits D5-D0 are ignored. The WDW function enables a user to set the WDS bit without concern that the watchdog timer value is modified. A logical diagram of the watchdog timer is shown in Figure 36 on page 22. Note that Page 21 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I setting the watchdog time out value to ‘0’ disables the watchdog function. The output of the watchdog timer is the flag bit WDF that is set if the watchdog is allowed to time out. If the watchdog interrupt enable (WIE) bit in the Interrupt register is set, a hardware interrupt on INT pin is also generated on watchdog timeout. The flag and the hardware interrupt are both cleared when user reads the flag registers. Figure 36. Watchdog Timer Block Diagram . Clock Divider Oscillator 32.768 KHz Zero Compare WDF Load Register WDS D Q WDW Q write to Watchdog Register The CY14X256I provides a backup power monitoring system that detects the backup power (either battery or capacitor backup) failure. The backup power fail flag (BPF) is issued on the next power-up in case of backup power failure. The BPF flag is set in the event of backup voltage falling lower than VBAKFAIL. The backup power is monitored even while the RTC is running in backup mode. Low voltage detected during backup mode is flagged through the BPF flag. BPF can hold the data only until a defined low level of the back up voltage (VDR). Interrupts 1 Hz 32 Hz Counter Backup Power Monitor Watchdog Register Programmable Square Wave Generator The square wave generator block uses the crystal output to generate a desired frequency on the INT pin of the device. The output frequency can be programmed to be one of the following: 1. 1 Hz 2. 512 Hz 3. 4096 Hz 4. 32768 Hz The CY14X256I has a flags register, interrupt register, and interrupt logic that can signal interrupt to the microcontroller. There are three potential sources for interrupt: watchdog timer, power monitor, and alarm timer. Each of these can be individually enabled to drive the INT pin by appropriate setting in the interrupt register (0x06). In addition, each has an associated flag bit in the flags register (0x00) that the host processor uses to determine the cause of the interrupt. The INT pin driver has two bits that specify its behavior when an interrupt occurs. An interrupt is raised only if both a flag is raised by one of the three sources and the respective interrupt enable bit in interrupts register is enabled (set to ‘1’). After an interrupt source is active, two programmable bits, H/L and P/L, determine the behavior of the output pin driver on INT pin. These two bits are located in the Interrupt register and can be used to drive level or pulse mode output from the INT pin. In pulse mode, the pulse width is internally fixed at approximately 200 ms. This mode is intended to reset a host microcontroller. In the level mode, the pin goes to its active polarity until the flags register is read by the user. This mode is used as an interrupt to a host microcontroller. The control bits are summarized in the following section. Interrupts are only generated while working on normal power and are not triggered when system is running in backup power mode. The square wave output is not generated while the device is running on backup power. Note CY14X256I generates valid interrupts only after the Powerup RECALL sequence is completed. All events on INT pin must be ignored for tFA duration after powerup. Power Monitor Interrupt Register The CY14X256I provides a power management scheme with power fail interrupt capability. It also controls the internal switch to back up power for the clock and protects the memory from low VCC access. The power monitor is based on an internal band gap reference circuit that compares the VCC voltage to VSWITCH threshold. Watchdog Interrupt Enable (WIE): When set to ‘1’, the watchdog timer drives the INT pin and an internal flag when a watchdog time out occurs. When WIE is set to ‘0’, the watchdog timer only affects the WDF flag in flags register. Alarm Interrupt Enable (AIE): When set to ‘1’, the alarm match drives the INT pin and an internal flag. When AIE is set to ‘0’, the alarm match only affects the AF flag in the flags register. Power Fail Interrupt Enable (PFE): When set to ‘1’, the power fail monitor drives the pin and an internal flag. When PFE is set to ‘0’, the power fail monitor only affects the PF flag in the flags register. Square Wave Enable (SQWE): When set to ‘1’, a square wave of programmable frequency is generated on the INT pin. The frequency is decided by the SQ1 and SQ0 bits of the interrupts register. This bit is nonvolatile and survives power cycle. The SQWE bit overrides all other interrupts. However, CAL bit will take precedence over the square wave generator. This bit defaults to ‘0’ from the factory. When VSWITCH is reached, as VCC decays from power loss, a data store operation is initiated from SRAM to the nonvolatile elements, securing the last SRAM data state. Power is also switched from VCC to the backup supply (battery or capacitor) to operate the RTC oscillator. When operating from the backup source, read and write operations to nvSRAM are inhibited and the RTC functions are not available to the user. The RTC clock continues to operate in the background. The updated RTC time keeping registers are available to the user after VCC is restored to the device (see nvSRAM Specifications on page 34). Document #: 001-65230 Rev. *B Page 22 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I High/Low (H/L): When set to a ‘1’, the INT pin is active HIGH and the driver mode is push pull. The INT pin drives HIGH only when VCC is greater than VSWITCH. When set to a ‘0’, the INT pin is active LOW and the drive mode is open drain. The INT pin must be pulled up to Vcc by a 10 k resistor while using the interrupt in active LOW mode. Pulse/Level (P/L): When set to a ‘1’ and an interrupt occurs, the INT pin is driven for approximately 200 ms. When P/L is set to a ‘0’, the INT pin is driven HIGH or LOW (determined by H/L) until the flags register is read. SQ1 and SQ0. These bits are used together to fix the frequency of square wave on INT pin output when SQWE bit is set to ‘1’. These bits are nonvolatile and survive power cycle. The output frequency is decided as illustrated in this table the flag and the pin. The pulse does not complete its specified duration if the flags register is read. If the INT pin is used as a host reset, the flags register is not read during a reset. Following is a summary table that shows the state of the INT pin, Table 9. State of the INT pin CAL SQWE WIE/AIE/ PFE INT Pin Output 1 X X 512 Hz 0 1 X Square wave output 0 0 1 Alarm 0 0 0 HI-Z Table 8. SQW Output Selection SQ1 SQ0 Frequency Comment 0 0 1 Hz 0 1 512 Hz 512 Hz clock output 1 0 4096 Hz 4 KHz clock output 1 1 32768 Hz Oscillator output frequency 1 Hz signal When an enabled interrupt source activates the INT pin, an external host reads the flag registers to determine the cause. Remember that all flags are cleared when the register is read. If the INT pin is programmed for Level mode, then the condition clears and the INT pin returns to its inactive state. If the pin is programmed for Pulse mode, then reading the flag also clears Document #: 001-65230 Rev. *B Flags Register The flags register has three flag bits: WDF, AF, and PF, which can be used to generate an interrupt. These flags are set by the watchdog timeout, alarm match, or power fail monitor respectively. The processor can either poll this register or enable interrupts to be informed when a flag is set. These flags are automatically reset after the register is read. The flags register is automatically loaded with the value 0x00 on power-up (except for the OSCF bit. See Stopping and Starting the Oscillator on page 20) Page 23 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 37. RTC Recommended Component Configuration Recommended Values C1 Y1 C2 Y1 = 32.768 KHz (12.5 pF) C1 = 12 pF C2 = 69 pF Xout Note: The recommended values for C1 and C2 include board trace capacitance. Xin Figure 38. Interrupt Block Diagram WIE Watchdog Timer WDF Power Monitor PFE PF AIE P/L 512 Hz Clock AF Pin Driver Mux Clock Alarm Square Wave HI-Z Control SEL Line VCC INT H/L VSS WDF - Watchdog timer flag WIE - Watchdog interrupt enable PF - Power fail flag PFE - Power Fail Enable AF - Alarm fag AIE - Alarm interrupt enable P/L - Pulse level H/L - High/Low SQWE - Square wave enable SQWE Priority CAL Encoder WIE/PIE/ AIE Document #: 001-65230 Rev. *B Page 24 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Table 10. RTC Register Map[2, 3] Register BCD Format Data D7 0x0F 0x0E D6 D5 D4 D3 D2 D1 10s years 0 0 0x0D 0 0 0x0C 0 0 0x0B 0 0 0x0A 0 0 10s months 10s day of month 0 0 D0 Years Years: 00–99 Months Months: 01–12 Day of month Day of month: 01–31 0 Day of week 10s hours 10s minutes Day of week: 01–07 Hours Hours: 00–23 Minutes Minutes: 00–59 0x09 0 0x08 OSCEN (0) 0x07 WDS (0) WDW (0) 0x06 WIE (0) AIE (0) 0x05 M (1) 0 10s alarm date Alarm day Alarm, day of month: 01–31 0x04 M (1) 0 10s alarm hours Alarm hours Alarm, hours: 00–23 0x03 M (1) 10s alarm minutes Alarm minutes Alarm, minutes: 00–59 0x02 M (1) 10s alarm seconds Alarm seconds Alarm, seconds: 00–59 0x01 0x00 10s seconds Function/Range 0 Seconds Cal Sign (0) AF Watchdog [4] WDT (000000) PFE (0) SQWE (0) H/L (1) 10s centuries WDF Seconds: 00–59 Calibration Values [4] Calibration (00000) PF P/L (0) SQ1 (0) SQ0 (0) Centuries OSCF[5] BPF[5] CAL (0) W (0) Interrupts [4] Centuries: 00–99 R (0) Flags [4] Notes 2. ( ) designates values shipped from the factory. 3. The unused bits of RTC registers are reserved for future use and should be set to ‘0’. 4. This is a binary value, not a BCD value. 5. When user resets OSCF and BPF flag bits, the flags register will be updated after tRTCp time. Document #: 001-65230 Rev. *B Page 25 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Table 11. Register Map Detail Register Description Time Keeping - Years D7 D6 0x0F D5 D4 D3 D2 10s years D1 D0 Years Contains the lower two BCD digits of the year. Lower nibble (four bits) contains the value for years; upper nibble (four bits) contains the value for 10s of years. Each nibble operates from 0 to 9. The range for the register is 0–99. Time Keeping - Months 0x0E D7 D6 D5 D4 0 0 0 10s month D3 D2 D1 D0 Months Contains the BCD digits of the month. Lower nibble (four bits) contains the lower digit and operates from 0 to 9; upper nibble (one bit) contains the upper digit and operates from 0 to 1. The range for the register is 1–12. Time Keeping - Date 0x0D D7 D6 0 0 D5 D4 D3 10s day of month D2 D1 D0 Day of month Contains the BCD digits for the date of the month. Lower nibble (four bits) contains the lower digit and operates from 0 to 9; upper nibble (two bits) contains the 10s digit and operates from 0 to 3. The range for the register is 1–31. Leap years are automatically adjusted for. Time Keeping - Day 0x0C D7 D6 D5 D4 D3 0 0 0 0 0 D2 D1 D0 Day of week Lower nibble (three bits) contains a value that correlates to day of the week. Day of the week is a ring counter that counts from 1 to 7 then returns to 1. The user must assign meaning to the day value, because the day is not integrated with the date. Time Keeping - Hours 0x0B D7 D6 0 0 D5 D4 D3 D2 10s hours D1 D0 Hours Contains the BCD value of hours in 24 hour format. Lower nibble (four bits) contains the lower digit and operates from 0 to 9; upper nibble (two bits) contains the upper digit and operates from 0 to 2. The range for the register is 0–23. Time Keeping - Minutes D7 0x0A D6 0 D5 D4 D3 D2 10s minutes D1 D0 Minutes Contains the BCD value of minutes. Lower nibble (four bits) contains the lower digit and operates from 0 to 9; upper nibble (three bits) contains the upper minutes digit and operates from 0 to 5. The range for the register is 0–59. Time Keeping - Seconds D7 0x09 D6 0 D5 D4 D3 D2 10s seconds D1 D0 Seconds Contains the BCD value of seconds. Lower nibble (four bits) contains the lower digit and operates from 0 to 9; upper nibble (three bits) contains the upper digit and operates from 0 to 5. The range for the register is 0–59. Calibration/Control 0X08 OSCEN D7 D6 D5 OSCEN 0 Calibration sign D4 D3 D2 D1 D0 Calibration Oscillator Enable. When set to ‘1’, the oscillator is stopped. When set to ‘0’, the oscillator runs. Disabling the oscillator saves battery or capacitor power during storage. Calibration Determines if the calibration adjustment is applied as an addition (1) to or as a subtraction (0) from the time-base. Sign Calibration These five bits control the calibration of the clock. Document #: 001-65230 Rev. *B Page 26 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Table 11. Register Map Detail (continued) Register Description Watchdog Timer 0x07 D7 D6 WDS WDW D5 D4 D3 D2 D1 D0 WDT WDS Watchdog Strobe. Setting this bit to ‘1’ reloads and restarts the watchdog timer. Setting the bit to ‘0’ has no effect. The bit is cleared automatically after the watchdog timer is reset. The WDS bit is write only. Reading it always returns a 0. WDW Watchdog Write Enable. Setting this bit to ‘1’ disables any WRITE to the watchdog timeout value (D5–D0). This enables the user to set the watchdog strobe bit without disturbing the timeout value. Setting this bit to ‘0’ allows bits D5–D0 to be written to the watchdog register when the next write cycle is complete. This function is explained in more detail in Watchdog Timer on page 21. WDT Watchdog timeout selection. The watchdog timer interval is selected by the 6-bit value in this register. It represents a multiplier of the 32 Hz count (31.25 ms). The range of timeout value is 31.25 ms (a setting of 1) to 2 seconds (setting of 3 Fh). Setting the watchdog timer register to ‘0’ disables the timer. These bits can be written only if the WDW bit was set to 0 on a previous cycle. Interrupt Status/Control 0x06 D7 D6 D5 D4 D3 D2 D1 D0 WIE AIE PFE SQWE H/L P/L SQ1 SQ0 WIE Watchdog Interrupt Enable. When set to ‘1’ and a watchdog timeout occurs, the watchdog timer drives the INT pin and the WDF flag. When set to ‘0’, the watchdog timeout affects only the WDF flag. AIE Alarm Interrupt Enable. When set to ‘1’, the alarm match drives the INT pin and the AF flag. When set to ‘0’, the alarm match only affects the AF flag. PFE Power Fail Enable. When set to ‘1’, the alarm match drives the INT pin and the PF flag. When set to ‘0’, the power fail monitor affects only the PF flag. SQWE Square Wave Enable. When set to ‘1’, a square wave is driven on the INT pin with frequency programmed using SQ1 and SQ0 bits. The square wave output takes precedence over interrupt logic. If the SQWE bit is set to ‘1’. when an enabled interrupt source becomes active, only the corresponding flag is raised and the INT pin continues to drive the square wave. H/L High/Low. When set to ‘1’, the INT pin is driven active HIGH. When set to ‘0’, the INT pin is open drain, active LOW. P/L Pulse/Level. When set to ‘1’, the INT pin is driven active (determined by H/L) by an interrupt source for approximately 200 ms. When set to ‘0’, the INT pin is driven to an active level (as set by H/L) until the flags register is read. SQ1, SQ0 SQ1, SQ0. These bits are used to decide the frequency of the Square wave on the INT pin output when SQWE bit is set to ‘1’. The following is the frequency output for each combination of (SQ1, SQ0): (0, 0) - 1 Hz (0, 1) - 512 Hz (1, 0) - 4096 Hz (1, 1) - 32768 Hz Alarm - Day 0x05 D7 D6 M 0 D5 D4 D3 D2 10s alarm date D1 D0 Alarm date Contains the alarm value for the date of the month and the mask bit to select or deselect the date value. M Match. When this bit is set to ‘0’, the date value is used in the alarm match. Setting this bit to ‘1’ causes the match circuit to ignore the date value. Alarm - Hours 0x04 D7 D6 M 0 D5 D4 10s alarm hours D3 D2 D1 D0 Alarm hours Contains the alarm value for the hours and the mask bit to select or deselect the hours value. M Match. When this bit is set to ‘0’, the hours value is used in the alarm match. Setting this bit to ‘1’ causes the match circuit to ignore the hours value. Document #: 001-65230 Rev. *B Page 27 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Table 11. Register Map Detail (continued) Register Description Alarm - Minutes 0x03 D7 D6 M D5 D4 D3 10s alarm minutes D2 D1 D0 Alarm minutes Contains the alarm value for the minutes and the mask bit to select or deselect the minutes value. M Match. When this bit is set to ‘0’, the minutes value is used in the alarm match. Setting this bit to ‘1’ causes the match circuit to ignore the minutes value. Alarm - Seconds 0x02 D7 D6 M D5 D4 D3 10s alarm seconds D2 D1 D0 Alarm seconds Contains the alarm value for the seconds and the mask bit to select or deselect the seconds’ value. M Match. When this bit is set to ‘0’, the seconds value is used in the alarm match. Setting this bit to ‘1’ causes the match circuit to ignore the seconds value. Time Keeping - Centuries 0x01 D7 D6 D5 D4 D3 D2 10s centuries D1 D0 Centuries Contains the BCD value of centuries. Lower nibble contains the lower digit and operates from 0 to 9; upper nibble contains the upper digit and operates from 0 to 9. The range for the register is 0-99 centuries. Flags 0x00 D7 D6 D5 D4 D3 D2 D1 D0 WDF AF PF OSCF BPF CAL W R WDF Watchdog Timer Flag. This read only bit is set to ‘1’ when the watchdog timer is allowed to reach ‘0’ without being reset by the user. It is cleared to ‘0’ when the flags register is read or on power-up AF Alarm Flag. This read only bit is set to ‘1’ when the time and date match the values stored in the alarm registers with the match bits = ‘0’. It is cleared when the flags register is read or on power-up. PF Power Fail Flag. This read only bit is set to ‘1’ when power falls below the power fail threshold VSWITCH. It is cleared when the flags register is read. OSCF Oscillator Fail Flag. Set to ‘1’ on power-up if the oscillator is enabled and not running in the first 5 ms of operation. This indicates that RTC backup power failed and clock value is no longer valid. This bit survives power cycle and is never cleared internally by the chip. The user must check for this condition and write '0' to clear this flag. When user resets OSCF flag bit, the bit will be updated after tRTCp time. BPF Backup Power Fail Flag. Set to ‘1’ on power-up if the backup power (battery or capacitor) failed. The backup power fail condition is determined by the voltage falling below their respective minimum specified voltage. BPF can hold the data only till a defined low level of the back up voltage (VDR). User must reset this bit to clear this flag. When user resets BPF flag bit, the bit will be updated after tRTCp time. CAL Calibration Mode. When set to ‘1’, a 512 Hz square wave is output on the INT pin. When set to ‘0’, the INT pin resumes normal operation. This bit takes priority than SQ0/SQ1 and other functions. This bit defaults to ‘0’ (disabled) on power-up. W Write Enable: Setting the ‘W’ bit to ‘1’ freezes updates of the RTC registers. The user can then write to RTC registers, alarm registers, calibration register, interrupt register and flags register. Setting the ‘W’ bit to ‘0’ causes the contents of the RTC registers to be transferred to the time keeping counters if the time has changed. This transfer process takes tRTCp time to complete. This bit defaults to 0 on power-up. R Read Enable: Setting ‘R’ bit to ‘1’, stops clock updates to user RTC registers so that clock updates are not seen during the reading process. Set ‘R’ bit to ‘0’ to resume clock updates to the holding register. Setting this bit does not require ‘W’ bit to be set to ‘1’. This bit defaults to ‘0’ on power-up. Document #: 001-65230 Rev. *B Page 28 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Best Practices ■ Power up boot firmware routines should rewrite the nvSRAM into the desired state (for example, AutoStore enabled). While the nvSRAM is shipped in a preset state, best practice is to again rewrite the nvSRAM into the desired state as a safeguard against events that might flip the bit inadvertently such as program bugs and incoming inspection routines. ■ The VCAP value specified in this datasheet includes a minimum and a maximum value size. Best practice is to meet this requirement and not exceed the maximum VCAP value because the nvSRAM internal algorithm calculates VCAP charge and discharge time based on this max VCAP value. Customers that want to use a larger VCAP value to make sure there is extra store charge and store time should discuss their VCAP size selection. ■ When base time is updated, these updates are transferred to the time keeping registers when ‘W’ bit is set to ‘0’. This transfer takes tRTCp time to complete. It is recommended to initiate software STORE or Hardware STORE after tRTCp time to save the base time into nonvolatile memory. nvSRAM products have been used effectively for over 26 years. While ease-of-use is one of the product’s main system values, experience gained working with hundreds of applications has resulted in the following suggestions as best practices: ■ The nonvolatile cells in this nvSRAM product are delivered by Cypress with 0x00 written in all cells. Incoming inspection routines at customer or contract manufacturer’s sites sometimes reprogram these values. Final NV patterns are typically repeating patterns of AA, 55, 00, FF, A5, or 5A. End product’s firmware should not assume an NV array is in a set programmed state. Routines that check memory content values to determine first time system configuration, cold or warm boot status, and so on should always program a unique NV pattern (that is, complex 4-byte pattern of 46 E6 49 53 hex or more random bytes) as part of the final system manufacturing test to ensure these system routines work consistently. Document #: 001-65230 Rev. *B Page 29 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Maximum Ratings Exceeding maximum ratings may shorten the useful life of the device. These user guidelines are not tested. Storage temperature ................................ –65 °C to +150 °C Maximum accumulated storage time Transient voltage (< 20 ns) on any pin to ground potential .................. –2.0 V to VCC + 2.0 V Package power dissipation capability (TA = 25 °C) .................................................. 1.0 W Surface mount lead soldering temperature (3 seconds) .......................................... +260 °C At 150 °C ambient temperature ....................... 1000 h DC output current (1 output at a time, 1s duration) ..... 15 mA At 85 °C ambient temperature ..................... 20 Years Static discharge voltage.......................................... > 2001 V (per MIL-STD-883, Method 3015) Ambient temperature with power applied ........................................... –55 °C to +150 °C Latch up current..................................................... > 140 mA Supply voltage on VCC relative to VSS CY14C256I: VCC = 2.4 V to 2.6 V......–0.5 V to +3.1 V CY14B256I: VCC = 2.7 V to 3.6 V ......–0.5 V to +4.1 V CY14E256I: VCC = 4.5 V to 5.5 V ......–0.5 V to +7.0 V DC voltage applied to outputs in high Z state ...................................... –0.5 V to VCC + 0.5 V Operating Range CY14B256I 2.7 V to 3.6 V Input voltage ........................................ –0.5 V to VCC + 0.5 V CY14E256I 4.5 V to 5.5 V Product CY14C256I Range Ambient Temperature VCC Industrial –40 °C to +85 °C 2.4 V to 2.6 V DC Electrical Characteristics Over the Operating Range Parameter Description VCC Test Conditions Power supply Min Typ[6] Max Unit CY14C256I 2.4 2.5 2.6 V CY14B256I 2.7 3.0 3.6 V CY14E256I 4.5 5.0 5.5 V ICC1 Average VCC current fSCL = 3.4 MHz; Values obtained without output loads (IOUT = 0 mA) – – 1 mA ICC2 Average VCC current during STORE All inputs don’t care, VCC = max Average current for duration tSTORE – – 2 mA ICC3 Average VCC current fSCL = 100 kHz; VCC = VCC (Typ), 25 °C All inputs cycling at CMOS levels. Values obtained without output loads (IOUT = 0 mA) – – 1 mA ICC4 Average VCAP current during AutoStore cycle All inputs don't care. Average current for duration tSTORE – – 3 mA ISB VCC standby current SCL > (VCC – 0.2 V). VIN < 0.2 V or > (VCC – 0.2 V). ‘W’ bit set to ‘0’. Standby current level after nonvolatile cycle is complete. Inputs are static. fSCL = 0 MHz. – – 250 A IZZ Sleep mode current tSLEEP time after SLEEP Instruction is registered. All inputs are static and configured at CMOS logic level. – – 8 A IIX[7] Input current in each I/O pin 0.1 VCC < Vi < 0.9 VCCmax (except HSB) –1 – +1 A –100 – +1 A –1 – +1 A Input current in each I/O pin (for HSB) IOZ Output leakage current Note 6. Typical values are at 25 °C, VCC = VCC (Typ). Not 100% tested. 7. Not applicable to WP, A2, A1 and A0 pins. Document #: 001-65230 Rev. *B Page 30 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I DC Electrical Characteristics (continued) Over the Operating Range Parameter Description Test Conditions Min Typ[6] Max Unit – – 7 pF Ci Capacitance for each I/O pin VIH Input HIGH voltage 0.7 Vcc – VCC + 0.5 V VIL Input LOW voltage – 0.5 – 0.3 VCC V V VOL Rin [8] Capacitance measured across all input and output signal pin and VSS. Output LOW voltage IOL = 3 mA 0 – 0.4 Input resistance (WP, A2, A1, A0) For VIN = VIL (Max) 50 – – K For VIN = VIH (Max) 1 – – M 0.05 VCC – – V CY14C256I 170 220 270 F CY14B256I CY14E256I 42 47 180 F Vhys Hysteresis of Schmitt trigger inputs VCAP Storage capacitor Between VCAP pin and VSS Data Retention and Endurance Parameter Description DATAR Data retention NVC Nonvolatile STORE operations Min Unit 20 Years 1,000 K 16-pin SOIC Unit 56.68 C/W 32.11 C/W Thermal Resistance Parameter[9] Description JA Thermal resistance (Junction to ambient) JC Thermal resistance (Junction to case) Test Conditions Test conditions follow standard test methods and procedures for measuring thermal impedance, according to EIA / JESD51. Notes 8. The input pull-down circuit is stronger (50 K) when the input voltage is below VIL and weak (1 M) when the input voltage is above VIH. 9. These parameters are guaranteed by design and are not tested. Document #: 001-65230 Rev. *B Page 31 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Figure 39. AC Test Loads and Waveforms For 3.0 V (CY14B256I) For 2.5 V (CY14C256I) For 5.0 V (CY14E256I) 3.0 V 2.5 V 5.0 V 867 700 1.6 K OUTPUT OUTPUT OUTPUT 100 pF 100 pF 50 pF AC Test Conditions Description CY14C256I CY14B256I CY14E256I 0 V to 2.5 V 0 V to 3 V 0 V to 5 V Input rise and fall times (10% - 90%) 10 ns 10 ns 10 ns Input and output timing reference levels 1.25 V 1.5 V 2.5 V Input pulse levels RTC Characteristics Parameter Description Min Typ Max Units 1.8 – 3.6 V – 0.45 0.6 µA 1.6 – 3.6 V VRTCbat RTC battery pin voltage IBAK[10] RTC backup current VRTCcap[11] RTC capacitor pin voltage VBAKFAIL Backup failure threshold 1.8 – 2 V VDR BPF flag retention voltage 1.6 – – V tOCS RTC oscillator time to start – 1 2 sec tRTCp RTC processing time from end of ‘W’ bit set to ‘0’ – – 1 ms RBKCHG RTC backup capacitor charge current-limiting resistor 350 – 850 Notes 10. Current drawn from either VRTCcap or VRTCbat when VCC < VSWITCH. 11. If VRTCcap > 0.5 V or if no capacitor is connected to VRTCcap pin, the oscillator will start in tOCS time. If a backup capacitor is connected and VRTCcap < 0.5 V, the capacitor must be allowed to charge to 0.5 V for oscillator to start. Document #: 001-65230 Rev. *B Page 32 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I AC Switching Characteristics Parameter 3.4 MHz[12] Description 1 MHz[12] 400 kHz[12] Unit Min Max Min Max Min Max – 3400 – 1000 – 400 kHz fSCL Clock frequency, SCL tSU; STA Setup time for Repeated START condition 160 – 250 – 600 – ns tHD;STA Hold time for START condition 160 – 250 – 600 – ns tLOW LOW period of the SCL 160 – 500 – 1300 – ns tHIGH HIGH period of the SCL 60 – 260 – 600 – ns tSU;DATA Data in setup time 10 – 100 tHD;DATA Data hold time (In/Out) 0 – 0 – 0 – ns tDH Data out hold time 0 – 0 – 0 – ns Rise time of SDA and SCL – 80 – 120 – 300 ns tr [13] tf[13] Fall time of SDA and SCL tSU;STO Setup time for STOP condition tVD;DATA tVD;ACK tOF [13] 100 ns – 80 – 120 – 300 ns 160 – 250 – 600 – ns Data output valid time – 130 – 400 – 900 ns ACK output valid time – 130 – 400 – 900 ns – 80 Output fall time from VIH min to VILmax tBUF Bus free time between STOP and next START condition 0.3 tSP Pulse width of spikes that must be suppressed by input filter – 5 – 120 – 300 ns 0.5 – 1.3 – us – 50 – 50 ns t SP tf ~ ~ ~ ~ Figure 40. Timing Diagram tr t LOW ~ ~ ~ ~ SDA t HD;STA t SU;DATA t BUF t HD;DATA t HIGH ~ ~ t HD;STA S ~ ~ SCL t SU;STA Sr tr tf t SU;STO P S Note 12. (Bus Load Capacitance (Cb) Considerations; Cb < 500 pF for I2C clock frequency (SCL) 100/400/1000 KHz; Cb < 100 pF for SCL at 3.4 MHz). 13. These parameters are guaranteed by design and are not tested. Document #: 001-65230 Rev. *B Page 33 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I nvSRAM Specifications Parameter tFA [14] Power-up RECALL duration tSTORE [15] tDELAY[16] tVCCRISE[17] VSWITCH STORE cycle duration Time allowed to complete SRAM write cycle VCC rise time Low voltage trigger level tLZHSB[17] HSB high to nvSRAM active time HSB output disable voltage VHDIS[17] tHHHD[17] tWAKE tSLEEP tSB Description CY14C256I CY14B256I CY14E256I CY14C256I CY14B256I CY14E256I HSB HIGH active time Time for nvSRAM to wake up from SLEEP mode Min – – – – – 150 – – – – – Max 40 20 20 8 25 – 2.35 2.65 4.40 5 1.9 Unit ms ms ms ms ns µs V V V µs V – – – – – – 500 40 20 20 8 100 ns ms ms ms ms µs CY14C256I CY14B256I CY14E256I Time to enter low power mode after issuing SLEEP instruction Time to enter into standby mode after issuing STOP condition Figure 41. AutoStore or Power-Up RECALL[18] VCC VSWITCH VHDIS t VCCRISE tHHHD Note15 15 tSTORE Note tHHHD 19 Note tSTORE 19 Note HSB OUT tDELAY tLZHSB AutoStore tLZHSB tDELAY POWERUP RECALL tFA tFA Read & Write Inhibited (RWI) POWER-UP RECALL Read & Write BROWN OUT AutoStore POWER-UP RECALL Read & Write POWER DOWN AutoStore Notes 14. tFA starts from the time VCC rises above VSWITCH. 15. If an SRAM write has not taken place since the last nonvolatile cycle, no AutoStore or Hardware STORE takes place. 16. On a Hardware STORE and AutoStore initiation, SRAM write operation continues to be enabled for time tDELAY. 17. These parameters are guaranteed by design and are not tested. 18. Read and Write cycles are ignored during STORE, RECALL, and while VCC is below VSWITCH. 19. During power-up and power-down, HSB glitches when HSB pin is pulled up through an external resistor. Document #: 001-65230 Rev. *B Page 34 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Software Controlled STORE/RECALL Cycles Parameter CY14X256I Description Min Max Unit tRECALL RECALL duration – 600 µs tSS[20, 21] Software sequence processing time – 500 µs Figure 42. Software STORE/RECALL Cycle[21] DATA OUTPUT BY MASTER Command Reg Address nvSRAM Control Slave Address acknowledge (A) by Slave acknowledge (A) by Slave SCL FROM MASTER 1 2 8 9 1 Command Byte (STORE/RECALL) 2 8 9 1 acknowledge (A) by Slave 2 8 9 S P START condition STOP condition RWI t STORE / t RECALL Figure 43. AutoStore Enable/Disable Cycle DATA OUTPUT BY MASTER Command Reg Address nvSRAM Control Slave Address acknowledge (A) by Slave acknowledge (A) by Slave SCL FROM MASTER 1 2 S START condition 8 9 1 Command Byte (ASENB/ASDISB) 2 8 9 1 acknowledge (A) by Slave 2 8 9 P STOP condition RWI t SS Notes 20. This is the amount of time it takes to take action on a soft sequence command. Vcc power must remain HIGH to effectively register command. 21. Commands such as STORE and RECALL lock out I/O until operation is complete which further increases this time. See the specific command. Document #: 001-65230 Rev. *B Page 35 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Hardware STORE Cycle Parameter tPHSB CY14X256I Description Min Max 15 – Hardware STORE pulse width Unit ns Figure 44. Hardware STORE Cycle[22] Write Latch set ~ ~ tPHSB HSB (IN) tSTORE tHHHD ~ ~ tDELAY HSB (OUT) tLZHSB RWI tPHSB HSB (IN) HSB pin is driven HIGH to VCC only by Internal 100 K: resistor, HSB driver is disabled SRAM is disabled as long as HSB (IN) is driven LOW. tDELAY RWI ~ ~ HSB (OUT) ~ ~ Write Latch not set Note 22. If an SRAM write has not taken place since the last nonvolatile cycle, no AutoStore or Hardware STORE takes place. Document #: 001-65230 Rev. *B Page 36 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Ordering Information Ordering Code Package Diagram Package Type Operating Range 51-85022 16-pin SOIC Industrial CY14B256I-SFXI CY14B256I-SFXIT The above part is Pb-free. This table contains Final information. Contact your local Cypress sales representative for availability of these parts. Ordering Code Definitions CY 14 B 256 I - SF X I T Option: T - Tape and Reel Blank - Std. Temperature: I - Industrial (–40 to 85 °C) Pb-free Package: SF - 16-pin SOIC I - Serial (I2C) nvSRAM with RTC Density: Voltage: C - 2.5 V B - 3.0 V E - 5.0 V 256 - 256 Kb 14 - nvSRAM Cypress Document #: 001-65230 Rev. *B Page 37 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Package Diagram Figure 45. 16-pin (300 mil) SOIC, 51-85022 51-85022 *C Document #: 001-65230 Rev. *B Page 38 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Acronyms Document Conventions Acronym Description Units of Measure BCD Binary coded decimal CMOS Complementary metal oxide semiconductor °C degrees Celsius CRC Cyclic redundancy check Hz Hertz EIA Electronic industries alliance I2C kbit 1024 bits Inter-integrated circuit bus kHz kilo Hertz I/O Input/output K kilo ohms JEDEC Joint Electron Devices Engineering Council A micro Amperes nvSRAM nonvolatile static random access memory mA milli Amperes OSCF Oscillator Fail Flag F micro Farads RoHS Restriction of hazardous substances Mbit/s Megabit per second R/W read/write MHz Mega Hertz RWI Read and write inhibited µs micro seconds SCL Serial clock line ms milli seconds SDA Serial data line ns nano seconds SNL serial number lock pF pico Farads SOIC Small outline integrated circuit V Volts ohms W Watts Document #: 001-65230 Rev. *B Symbol Unit of Measure Page 39 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Document History Page Document Title: CY14C256I, CY14B256I, CY14E256I 256-Kbit (32 K × 8) Serial (I2C) nvSRAM with Real Time Clock Document Number: 001-65230 Rev. ECN No. Submission Date Orig. of Change ** 3089600 11/18/2010 GVCH New datasheet. *A 3197569 03/16/2011 GVCH Updated AutoStore Operation (description). Updated Hardware STORE and HSB pin Operation (Added more clarity on HSB pin operation). Updated Table 6 (Product ID column). Updated Setting the Clock (description). Updated Figure 37 (C1, C2 values to 12pF, 69pF from 10pF, 67pF respectively). Updated Table 11 (‘W’ bit description). Updated Best Practices. Updated RTC Characteristics (Added tRTCp parameter). Updated nvSRAM Specifications (description of tLZHSB parameter). Fixed Typo error in Figure 41. Updated in new template. *B 3248510 05/05/2011 GVCH Datasheet status changed from “Preliminary” to “Final” Document #: 001-65230 Rev. *B Description of Change Page 40 of 41 [+] Feedback CY14C256I CY14B256I, CY14E256I Sales, Solutions, and Legal Information Worldwide Sales and Design Support Cypress maintains a worldwide network of offices, solution centers, manufacturer’s representatives, and distributors. To find the office closest to you, visit us at Cypress Locations. Products Automotive Clocks & Buffers Interface Lighting & Power Control PSoC Solutions cypress.com/go/automotive cypress.com/go/clocks psoc.cypress.com/solutions cypress.com/go/interface PSoC 1 | PSoC 3 | PSoC 5 cypress.com/go/powerpsoc cypress.com/go/plc Memory Optical & Image Sensing PSoC Touch Sensing USB Controllers Wireless/RF cypress.com/go/memory cypress.com/go/image cypress.com/go/psoc cypress.com/go/touch cypress.com/go/USB cypress.com/go/wireless © Cypress Semiconductor Corporation, 2010-2011. The information contained herein is subject to change without notice. Cypress Semiconductor Corporation assumes no responsibility for the use of any circuitry other than circuitry embodied in a Cypress product. Nor does it convey or imply any license under patent or other rights. Cypress products are not warranted nor intended to be used for medical, life support, life saving, critical control or safety applications, unless pursuant to an express written agreement with Cypress. Furthermore, Cypress does not authorize its products for use as critical components in life-support systems where a malfunction or failure may reasonably be expected to result in significant injury to the user. The inclusion of Cypress products in life-support systems application implies that the manufacturer assumes all risk of such use and in doing so indemnifies Cypress against all charges. Any Source Code (software and/or firmware) is owned by Cypress Semiconductor Corporation (Cypress) and is protected by and subject to worldwide patent protection (United States and foreign), United States copyright laws and international treaty provisions. Cypress hereby grants to licensee a personal, non-exclusive, non-transferable license to copy, use, modify, create derivative works of, and compile the Cypress Source Code and derivative works for the sole purpose of creating custom software and or firmware in support of licensee product to be used only in conjunction with a Cypress integrated circuit as specified in the applicable agreement. Any reproduction, modification, translation, compilation, or representation of this Source Code except as specified above is prohibited without the express written permission of Cypress. Disclaimer: CYPRESS MAKES NO WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, WITH REGARD TO THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. Cypress reserves the right to make changes without further notice to the materials described herein. Cypress does not assume any liability arising out of the application or use of any product or circuit described herein. Cypress does not authorize its products for use as critical components in life-support systems where a malfunction or failure may reasonably be expected to result in significant injury to the user. The inclusion of Cypress’ product in a life-support systems application implies that the manufacturer assumes all risk of such use and in doing so indemnifies Cypress against all charges. Use may be limited by and subject to the applicable Cypress software license agreement. Document #: 001-65230 Rev. *B Revised May 5, 2011 Page 41 of 41 All products and company names mentioned in this document may be the trademarks of their respective holders. [+] Feedback