Please note that Cypress is an Infineon Technologies Company.
The document following this cover page is marked as “Cypress” document as this is the
company that originally developed the product. Please note that Infineon will continue
to offer the product to new and existing customers as part of the Infineon product
portfolio.
Continuity of document content
The fact that Infineon offers the following product as part of the Infineon product
portfolio does not lead to any changes to this document. Future revisions will occur
when appropriate, and any changes will be set out on the document history page.
Continuity of ordering part numbers
Infineon continues to support existing part numbers. Please continue to use the
ordering part numbers listed in the datasheet for ordering.
www.infineon.com
FM31L276/FM31L278
64-Kbit/256-Kbit Integrated Processor
Companion with F-RAM
64-Kbit/256-Kbit Integrated Processor Companion with F-RAM
Features
■
■
■
■
■
64-Kbit/256-Kbit ferroelectric random access memory (F-RAM)
❐ Logically organized as 8K × 8 (FM31L276)/
32K × 8 (FM31L278)
14
❐ High-endurance 100 trillion (10 ) read/writes
❐ 151-year data retention (See Data Retention and Endurance
on page 29)
❐ NoDelay™ writes
❐ Advanced high-reliability ferroelectric process
High Integration Device Replaces Multiple Parts
❐ Serial nonvolatile memory
❐ Real time clock (RTC)
❐ Low voltage reset
❐ Watchdog timer
❐ Early power-fail warning/NMI
❐ Two 16-bit event counter
❐ Serial number with write-lock for security
Real-time Clock/Calendar
❐ Backup current at 2 V: 1.15 A at +25 C
❐ Seconds through centuries in BCD format
❐ Tracks leap years through 2099
❐ Uses standard 32.768 kHz crystal (6 pF/12.5 pF)
❐ Software calibration
❐ Supports battery or capacitor backup
Processor Companion
❐ Active-low reset output for VDD and watchdog
❐ Programmable low-VDD reset trip point
❐ Manual reset filtered and debounced
❐ Programmable watchdog timer
❐ Dual Battery-backed event counter tracks system intrusions
or other events
❐ Comparator for power-fail interrupt
❐ 64-bit programmable serial number with lock
Fast 2-wire serial interface (I2C)
❐ Up to 1-MHz frequency
❐ Supports legacy timings for 100 kHz and 400 kHz
2
❐ RTC, Supervisor controlled via I C interface
❐ Device select pins for up to 4 memory devices
■
Low power consumption
❐ 1.5 mA active current at 1 MHz
❐ 120 A standby current
■
Operating voltage: VDD = 2.7 V to 3.6 V
■
Industrial temperature: –40 C to +85 C
■
14-pin small outline integrated circuit (SOIC) package
Cypress Semiconductor Corporation
Document Number: 001-86392 Rev. *D
■
Restriction of hazardous substances (RoHS) compliant
■
Underwriters laboratory (UL) recognized
Functional Description
The FM31L276/FM31L278 device integrates F-RAM memory
with the most commonly needed functions for processor-based
systems. Major features include nonvolatile memory, real time
clock, low-VDD reset, watchdog timer, nonvolatile event counter,
lockable 64-bit serial number area, and general purpose
comparator that can be used for a power-fail (NMI) interrupt or
any other purpose.
The FM31L276/FM31L278 is a 64-Kbit/256-Kbit nonvolatile
memory employing an advanced ferroelectric process. A
ferroelectric random access memory or F-RAM is nonvolatile
and performs reads and writes similar to a RAM. This memory is
truly nonvolatile rather than battery backed. It provides reliable
data retention for 151 years while eliminating the complexities,
overhead, and system-level reliability problems caused by other
nonvolatile memories. The FM31L276/FM31L278 is capable of
supporting 1014 read/write cycles, or 100 million times more write
cycles than EEPROM.
The real time clock (RTC) provides time and date information in
BCD format. It can be permanently powered from an external
backup voltage source, either a battery or a capacitor. The
timekeeper uses a common external 32.768 kHz crystal and
provides a calibration mode that allows software adjustment of
timekeeping accuracy.
The processor companion includes commonly needed CPU
support functions. Supervisory functions include a reset output
signal controlled by either a low VDD condition or a watchdog
timeout. RST goes active when VDD drops below a
programmable threshold and remains active for 100 ms after
VDD rises above the trip point. A programmable watchdog timer
runs from 100 ms to 3 seconds. The watchdog timer is optional,
but if enabled it will assert the reset signal for 100 ms if not
restarted by the host before the timeout. A flag-bit indicates the
source of the reset.
A comparator on PFI compares an external input pin to the
onboard 1.2 V reference. This is useful for generating a
power-fail interrupt (NMI) but can be used for any purpose. The
family also includes a programmable 64-bit serial number that
can be locked making it unalterable. Additionally it offers a dual
battery-backed event counter that tracks the number of rising or
falling edges detected on a dedicated input pin.
For a complete list of related documentation, click here.
•
198 Champion Court
•
San Jose, CA 95134-1709
•
408-943-2600
Revised November 19, 2018
FM31L276/FM31L278
Logic Block Diagram
Document Number: 001-86392 Rev. *D
Page 2 of 36
FM31L276/FM31L278
Contents
Pinout ................................................................................ 4
Pin Definitions .................................................................. 4
Functional Overview ........................................................ 5
Memory Architecture ................................................... 5
Processor Companion ..................................................... 5
Processor Supervisor .................................................. 5
Manual Reset .............................................................. 6
Reset Flags ................................................................. 6
Early Power Fail Comparator ...................................... 7
Event Counter ............................................................. 7
Serial Number ............................................................. 7
Real-time Clock Operation ............................................... 8
Backup Power ............................................................. 9
Trickle Charger ............................................................ 9
Calibration ................................................................... 9
Crystal Oscillator ....................................................... 10
Layout Recommendations ............................................. 11
Register Map ................................................................... 14
I2C Interface .................................................................... 21
STOP Condition (P) ................................................... 21
START Condition (S) ................................................. 21
Data/Address Transfer .............................................. 21
Acknowledge/No-acknowledge ................................. 22
Slave Address ........................................................... 22
Addressing Overview - Memory ................................ 23
Addressing Overview - RTC & Companion ............... 23
Data Transfer ............................................................ 23
Memory Operation .......................................................... 24
Memory Write Operation ........................................... 24
Document Number: 001-86392 Rev. *D
Memory Read Operation ........................................... 25
RTC/Companion Write Operation ............................. 26
RTC/Companion Read Operation ............................. 26
Addressing FRAM Array
in the FM31L276/FM31L278 Family ................................. 26
Maximum Ratings ........................................................... 27
Operating Range ............................................................. 27
DC Electrical Characteristics ........................................ 27
Data Retention and Endurance ..................................... 29
Capacitance .................................................................... 29
Thermal Resistance ........................................................ 29
AC Test Loads and Waveforms ..................................... 29
AC Test Conditions ........................................................ 29
Supervisor Timing .......................................................... 30
AC Switching Characteristics ....................................... 31
Ordering Information ...................................................... 32
Ordering Code Definitions ......................................... 32
Package Diagram ............................................................ 33
Acronyms ........................................................................ 34
Document Conventions ................................................. 34
Units of Measure ....................................................... 34
Document History Page ................................................. 35
Sales, Solutions, and Legal Information ...................... 36
Worldwide Sales and Design Support ....................... 36
Products .................................................................... 36
PSoC® Solutions ...................................................... 36
Cypress Developer Community ................................. 36
Technical Support ..................................................... 36
Page 3 of 36
FM31L276/FM31L278
Pinout
Figure 1. 14-pin SOIC pinout
CNT1
1
14
VDD
CNT2
2
13
SCL
A0
3
12
SDA
A1
4
11
X2
CAL/PFO
5
10
X1
RST
6
9
PFI
VSS
7
8
VBAK
Pin Definitions
Pin Name
I/O Type
Description
A1–A0
Input
Device Select Address 1–0. These pins are used to select one of up to 4 devices of the same type
on the same I2C bus. To select the device, the address value on the three pins must match the
corresponding bits contained in the slave address. The address pins are pulled down internally.
SDA
Input/Output Serial Data/Address. This is a bi-directional pin for the I2C interface. It is open-drain and is intended
to be wire-OR'd with other devices on the I2C bus. The input buffer incorporates a Schmitt trigger for
noise immunity and the output driver includes slope control for falling edges. An external pull-up
resistor is required.
SCL
Input
Serial Clock. The serial clock pin for the I2C interface. Data is clocked out of the device on the falling
edge, and into the device on the rising edge. The SCL input also incorporates a Schmitt trigger input
for noise immunity.
CNT1, CNT2
Input
Event Counter Inputs. These battery-backed inputs increment counters when an edge is detected
on the corresponding CNT pin. The polarity is programmable. These pins should not be left floating.
Tie to ground if these pins are not used.
X1, X2
Input/Output 32.768 kHz crystal connection. When using an external oscillator, apply the clock to X1 and a DC
mid-level to X2. These pins should be left unconnected if RTC is not used.
RST
Input/Output Reset. This active-low output is open drain with weak pull-up. It is also an input when used as a manual
reset. This pin should be left floating if unused.
PFI
Input
Early Power-fail Input. Typically connected to an unregulated power supply to detect an early power
failure. This pin must be tied to ground if unused.
CAL/PFO
Output
Calibration/Early Power-fail Output. In calibration mode, this pin supplies a 512 Hz square-wave
output for clock calibration. In normal operation, this is the early power-fail output.
VBAK
Power supply Backup supply voltage. Connected to a 3 V battery or a large value capacitor. If no backup supply
is used, this pin should be tied to ground and the VBC bit should be cleared in the RTC register 0Bh.
The trickle charger is UL recognized and ensures no excessive current when using a lithium battery.
VSS
Power supply Ground for the device. Must be connected to the ground of the system.
VDD
Power supply Power supply input to the device.
Document Number: 001-86392 Rev. *D
Page 4 of 36
FM31L276/FM31L278
Functional Overview
Processor Companion
The FM31L276/FM31L278 device combines a serial nonvolatile
RAM with a real time clock (RTC) and a processor companion.
The companion is a highly integrated peripheral including a
processor supervisor, a comparator used for early power-fail
warning, nonvolatile event counters, and a 64-bit serial number.
The FM31L276/FM31L278 integrates these complementary but
distinct functions under a common interface in a single package.
The product is organized as two logical devices. The first is a
memory and the second is the companion which includes all the
remaining functions. From the system perspective they appear
to be two separate devices with unique IDs on the serial bus.
In addition to nonvolatile RAM, the FM31L276/FM31L278
incorporates a real time clock and highly integrated processor
companion. The companion includes a low-VDD reset, a
programmable watchdog timer, a battery-backed event
counters, a comparator for early power-fail detection or other
purposes, and a 64-bit serial number.
The memory is organized as a standalone nonvolatile I2C
memory using standard device ID value. The real time clock and
supervisor functions are accessed with a separate I2C device ID.
This allows clock/calendar data to be read while maintaining the
most recently used memory address. The clock and supervisor
functions are controlled by 25 special function registers. The
RTC and event counter circuits are maintained by the power
source on the VBAK pin, allowing them to operate from battery or
backup capacitor power when VDD drops below a set threshold.
Each functional block is described below.
Memory Architecture
The FM31L276/FM31L278 device is available in memory size
64-Kbit/256-Kbit. The device uses two-byte addressing for the
memory portion of the chip. This makes the device software
compatible with its standalone memory counterparts, but makes
them compatible within the entire family.
The memory array is logically organized as 8,192 × 8 bits/
32,768 × 8 bits and is accessed using an industry-standard I2C
interface. The memory is based on F-RAM technology.
Therefore it can be treated as RAM and is read or written at the
speed of the I2C bus with no delays for write operations. It also
offers effectively unlimited write endurance unlike other
nonvolatile memory technologies. The I2C protocol is described
on I2C Interface on page 21.
The memory array can be write-protected by software. Two bits
in the processor companion area (WP1, WP0 in register 0Bh)
control the protection setting. Based on the setting, the protected
addresses cannot be written and the I2C interface will not
acknowledge any data to protected addresses. The special
function registers containing these bits are described in detail in
Table 1 below.
Processor Supervisor
Supervisors provide a host processor two basic functions:
detection of power supply fault conditions and a watchdog timer
to
escape
a
software
lockup
condition.
The
FM31L276/FM31L278 has a reset pin (RST) to drive a processor
reset input during power faults, power-up, and software lockups.
It is an open drain output with a weak internal pull-up to VDD. This
allows other reset sources to be wire-OR’d to the RST pin. When
VDD is above the programmed trip point, RST output is pulled
weakly to VDD. If VDD drops below the reset trip point voltage
level (VTP), the RST pin will be driven LOW. It will remain LOW
until VDD falls too low for circuit operation which is the VRST level.
When VDD rises again above VTP, RST continues to drive LOW
for at least 100 ms (tRPU) to ensure a robust system reset at a
reliable VDD level. After tRPU has been met, the RST pin will
return to the weak HIGH state. While RST is asserted, serial bus
activity is locked out even if a transaction occurred as VDD
dropped below VTP. A memory operation started while VDD is
above VTP will be completed internally.
Table 2 below shows how bit VTP controls the trip point of the
low-VDD reset. They are located in register 0Bh, bits 1 and 0. The
reset pin will drive LOW when VDD is below the selected VTP
voltage, and the I2C interface and F-RAM array will be locked
out. Note that the bit 1 location is a don't care. Figure 2 illustrates
the reset operation in response to a low VDD.
Table 2. VTP setting
VTP Setting
VTP
2.6 V
0
2.9 V
1
Figure 2. Low VDD Reset
VDD
VTP
tRPU
Table 1. Block Memory Write Protection
WP1
WP0
Protected Address Range
0
0
None
0
1
Bottom 1/4
1
0
Bottom 1/2
1
1
Full array
Document Number: 001-86392 Rev. *D
RST
Page 5 of 36
FM31L276/FM31L278
A watchdog timer can also be used to drive an active reset signal.
The watchdog is a free-running programmable timer. The
timeout period can be software programmed from 100 ms to
3 seconds in 100 ms increments via a 5-bit nonvolatile register.
All programmed settings are minimum values and vary with
temperature according to the operating specifications. The
watchdog has two additional controls associated with its
operation, a watchdog enable bit (WDE) and timer restart bits
(WR). Both the enable bit must be set and the watchdog must
timeout in order to drive RST active. If a reset event occurs, the
timer will automatically restart on the rising edge of the reset
pulse. If WDE = ‘0’, the watchdog timer runs but a watchdog fault
will not cause RST to be asserted LOW. The WTR flag will be
set, indicating a watchdog fault. This setting is useful during
software development if the developer does not want RST to
drive. Note that setting the maximum timeout setting (11111b)
disables the counter to save power. The second control is a
nibble that restarts the timer preventing a reset. The timer should
be restarted after changing the timeout value.
The watchdog timeout value is located in register 0Ah, bits 4:0,
and the watchdog enable is bit 7. The watchdog is restarted by
writing the pattern 1010b to the lower nibble of register 09h.
Writing this pattern will also cause the timer to load new timeout
values. Writing other patterns to this address will not affect its
operation. Note the watchdog timer is free-running. Prior to
enabling it, users should restart the timer as described above.
This assures that the full timeout period will be set immediately
after enabling. The watchdog is disabled when VDD is below VTP.
The following table summarizes the watchdog bits. A block
diagram follows.
Watchdog Timeout
Watchdog Enable
Watchdog Restart
WDT(4:0)
WDE
WR(3:0)
0Ah, bits 4:0
0Ah, bit 7
09h, bits 3:0
Manual Reset
The RST is a bi-directional signal allowing the
FM31L276/FM31L278 to filter and de-bounce a manual reset
switch. The RST input detects an external low condition and
responds by driving the RST signal LOW for 100 ms.
Figure 4. Manual Reset
MCU
Reset
Switch
RST
RST
FM31L276/FM31L278
FM31L276/FM31L278
drives
100 ms (min.)
Note The internal weak pull-up eliminates the need for additional
external components.
Reset Flags
In case of a reset condition, a flag bit will be set to indicate the
source of the reset. A low-VDD reset is indicated by the POR flag,
register 09h bit 6. A watchdog reset is indicated by the WTR flag,
register 09h bit 7. Note that the flags are internally set in
response to reset sources, but they must be cleared by the user.
When the register is read, it is possible that both flags are set if
both have occurred since the user last cleared them.
Figure 3. Watchdog Timer
Timebase
100 ms
clock
WR(3:0) = 1010b to restart
Down Counter
Watchdog
Timer Settings
Document Number: 001-86392 Rev. *D
RST
WDE
Page 6 of 36
FM31L276/FM31L278
Early Power Fail Comparator
An early power fail warning can be provided to the processor well
before VDD drops out of spec. The comparator is used to create
a power fail interrupt (NMI). This can be accomplished by
connecting the PFI pin to the unregulated power supply via a
resistor divider. An application circuit is shown below.
Figure 5. Comparator as a Power-Fail Warning
Regulator
read anytime VDD is above VTP, and they will be incremented as
long as a valid VBAK power source is provided. To read, set the
RC bit register 0Ch bit 3 to 1. This takes a snapshot of all four
counter bytes allowing a stable value even if a count occurs
during the read. The registers can be written by software allowing
the counters to be cleared or initialized by the system. Counts
are blocked during a write operation. The two counters can be
cascaded to create a single 32-bit counter by setting the CC
control bit (register 0Ch, bit 2). When cascaded, the CNT1 input
will cause the counter to increment. CNT2 is not used in this
mode and should be tied to ground.
V DD
Figure 6. Event Counter
C1P
16-bit Counter
CNT1
FM31L276/
FM31L278
To MCU CAL/PFO
NMI input
+
-
C 2P
PFI
CNT2
16-bit Counter
1.2 V ref
CC
The voltage on the PFI input pin is compared to an onboard 1.2 V
reference. When the PFI input voltage drops below this
threshold, the comparator will drive the CAL/PFO pin to a LOW
state. The comparator has 100 mV (max) of hysteresis to reduce
noise sensitivity, only for a rising PFI signal. For a falling PFI
edge, there is no hysteresis.
The comparator is a general purpose device and its application
is not limited to the NMI function.
The comparator is not integrated into the special function
registers except as it shares its output pin with the CAL output.
When the RTC calibration mode is invoked by setting the CAL
bit (register 00h, bit 2), the CAL/PFO output pin will be driven with
a 512 Hz square wave and the comparator will be ignored. Since
most users only invoke the calibration mode during production,
this should have no impact on system operations using the
comparator.
Note The maximum voltage on the comparator input PFI is
limited to 3.75 V under normal operating conditions.
Event Counter
The FM31L276/FM31L278 offers the user two battery-backed
event counters. Input pins CNT1 and CNT2 are programmable
edge detectors. Each clocks a 16-bit counter. When an edge
occurs, the counters will increment their respective registers.
Counter 1 is located in registers 0Dh and 0Eh, Counter 2 is
located in registers 0Fh and 10h. These register values can be
Document Number: 001-86392 Rev. *D
The control bits for event counting are located in register 0Ch.
Counter 1 Polarity is bit C1P, bit 0; Counter 2 Polarity is C2P, bit
1; the Cascade Control is CC, bit 2; and the Read Counter bit is
RC, bit 3.
The polarity bits must be set prior to setting the counter value(s).
If a polarity bit is changed, the counter may inadvertently
increment. If the counter pins are not being used, tie them to
ground.
Serial Number
A memory location to write a 64-bit serial number is provided. It
is a writeable nonvolatile memory block that can be locked by the
user once the serial number is set. The 8 bytes of data and the
lock bit are all accessed via the device ID for the Processor
Companion. Therefore the serial number area is separate and
distinct from the memory array. The serial number registers can
be written an unlimited number of times, so these locations are
general purpose memory. However, once the lock bit is set, the
values cannot be altered and the lock cannot be removed. Once
locked the serial number registers can still be read by the
system.
The serial number is located in registers 11h to 18h. The lock bit
is SNL (register 0Bh, bit 7). Setting the SNL bit to a ‘1’ disables
writes to the serial number registers, and the SNL bit cannot be
cleared.
Page 7 of 36
FM31L276/FM31L278
Real-time Clock Operation
the R bit from ‘0’ to ‘1’ transfers timekeeping information from the
core into holding registers that can be read by the user. If a
timekeeper update is pending when R is set, then the core will
be updated prior to loading the user registers. The registers are
frozen and will not be updated again until the R bit is cleared to
‘0’. R is used for reading the time.
The real-time clock (RTC) is a timekeeping device that can be
battery or capacitor backed for permanently-powered operation.
It offers a software calibration feature that allows high accuracy.
The RTC consists of an oscillator, clock divider, and a register
system for user access. It divides down the 32.768 kHz
time-base and provides a minimum resolution of seconds (1 Hz).
Static registers provide the user with read/write access to the
time values. It includes registers for seconds, minutes, hours,
day-of-the-week, date, months, and years. A block diagram
(Figure 7) illustrates the RTC function.
Setting the W bit to ‘1’ locks the user registers. Clearing it to ‘0’
causes the values in the user registers to be loaded into the
timekeeper core. W bit is used for writing new time values. Users
should be certain not to load invalid values, such as FFh, to the
timekeeping registers. Updates to the timekeeping core occur
continuously except when locked.
The user registers are synchronized with the timekeeper core
using R and W bits in register 00h described below. Changing
Figure 7. Real-Time Clock Core Block Diagram
OSCEN
Oscillator
32.768 kHz
crystal
512 Hz or
Square Wave
Clock
CF
Years
8 bits
Months
5 bits
Date
6 bits
Days
3 bits
Hours
6 bits
User Interface Registers
Document Number: 001-86392 Rev. *D
Divider
1 Hz
MInutes
7 bits
W
Update
Logic
Seconds
7 bits
R
Page 8 of 36
FM31L276/FM31L278
Backup Power
Trickle Charger
The real-time clock/calendar is intended to be permanently
powered. When the primary system power fails, the voltage on
the VDD pin will drop. When VDD is less than 2.5 V, the RTC (and
event counters) will switch to the backup power supply on VBAK.
The clock operates at extremely low current in order to maximize
battery or capacitor life. However, an advantage of combining a
clock function with F-RAM memory is that data is not lost
regardless of the backup power source.
To facilitate capacitor backup the VBAK pin can optionally provide
a trickle charge current. When the VBC bit (register 0Bh, bit 2) is
set to ‘1’, the VBAK pin will source approximately 80 µA until VBAK
reaches VDD. This charges the capacitor to VDD without an
external diode and resistor charger. There is a Fast Charge
mode which is enabled by the FC bit (register 0Bh, bit 5). In this
mode the trickle charger current is set to approximately 1 mA,
allowing a large backup capacitor to charge more quickly.
The IBAK current varies with temperature and voltage (see DC
Electrical Characteristics on page 27). The following graph
shows IBAK as a function of VBAK. These curves are useful for
calculating backup time when a capacitor is used as the VBAK
source.
In the case where no battery is used, the VBAK pin should be tied
to VSS. Be sure to turn off the trickle charger (VBC = ‘0’),
otherwise charger current will be shunted to ground from VDD.
Figure 8. IBAK vs. VBAK Voltage
Note Systems using lithium batteries should clear the VBC bit to
‘0’ to prevent battery charging. The VBAK circuitry includes an
internal 1 K series resistor as a safety element. The trickle
charger is UL Recognized.
IBAK (A)
Calibration
VBAK (V)
The minimum VBAK voltage varies linearly with temperature. The
user can expect the minimum VBAK voltage to be 1.23 V at
+85 °C and 1.90 V at -40 °C. The tested limit is 1.55 V at +25 °C.
Note The minimum VBAK voltage has been characterized at
-40 °C and +85 °C but is not 100% tested.
The calibration setting is stored in F-RAM so it is not lost should
the backup source fail. It is accessed with bits CAL(4:0) in
register 01h. This value can be written only when the CAL bit is
set to a ‘1’. To exit the calibration mode, the user must clear the
CAL bit to a ‘0’. When the CAL bit is ‘0’, the CAL/PFO pin will
revert to the power fail output function.
VBAKmin. (V)
Figure 9. VBAK(min.) vs Temperature
When the CAL bit in the register 00h is set to ‘1’, the clock enters
calibration mode. In calibration mode, the CAL/PFO output pin is
dedicated to the calibration function and the power fail output is
temporarily unavailable. Calibration operates by applying a
digital correction to the counter based on the frequency error. In
this mode, the CAL/PFO pin is driven with a 512 Hz (nominal)
square wave. Any measured deviation from 512 Hz translates
into a timekeeping error. The user converts the measured error
in ppm and writes the appropriate correction value to the
calibration register. The correction factors are listed in the table
below. Positive ppm errors require a negative adjustment that
removes pulses. Negative ppm errors require a positive
correction that adds pulses. Positive ppm adjustments have the
CALS (sign) bit set to ‘1’, whereas negative ppm adjustments
have CALS = ‘0’. After calibration, the clock will have a maximum
error of ±2.17 ppm or ±0.09 minutes per month at the calibrated
temperature.
Temperature (°C)
Document Number: 001-86392 Rev. *D
Page 9 of 36
FM31L276/FM31L278
Crystal Oscillator
The crystal oscillator is designed to use a 6 pF/12.5 pF crystal
without the need for external components, such as loading
capacitors. The FM31L276/FM31L278 device has built-in
loading capacitors that are optimized for use with 6 pF crystals,
but which work well with 12.5 pF crystals. For either crystal, no
additional external loading capacitors are required nor
suggested.
In the example, R1 and R2 are chosen such that the X2 voltage
is centered around the X1 oscillator drive levels. If you wish to
avoid the DC current, you may choose to drive X1 with an
external clock and X2 with an inverted clock using a CMOS
inverter.
If a 32.768 kHz crystal is not used, an external oscillator may be
connected to the FM31L276/FM31L278. Apply the oscillator to
the X1 pin. Its high and low voltage levels can be driven rail-to-rail
or amplitudes as low as approximately 500 mV p-p. To ensure
proper operation, a DC bias must be applied to the X2 pin. It
should be centered between the high and low levels on the X1
pin. This can be accomplished with a voltage divider.
Figure 10. External Oscillator
FM31L276/FM31L278
X1
X2
VDD
R1
R2
Document Number: 001-86392 Rev. *D
Page 10 of 36
FM31L276/FM31L278
Layout Recommendations
The X1 and X2 crystal pins employ very high impedance circuits
and the oscillator connected to these pins can be upset by noise
or extra loading. To reduce RTC clock errors from signal
switching noise, a guard ring must be placed around these pads
and the guard ring grounded. SDA and SCL traces should be
routed away from the X1 / X2 pads. The X1 and X2 trace lengths
should be less than 5 mm. The use of a ground plane on the
backside or inner board layer is preferred. See layout example.
Red is the top layer, green is the bottom layer.
Figure 11. Layout Recommendations
VDD
VDD
SCL
SCL
SDA
SDA
X2
X2
X1
X1
PFI
PFI
VBAK
VBA K
Layout for Surface Mount Crystal
Layout for Through Hole Crystal
(red = top layer, green = bottom layer)
(red = top layer, green = bottom layer)
Document Number: 001-86392 Rev. *D
Page 11 of 36
FM31L276/FM31L278
Table 3. Digital Calibration Adjustments
S.No.
Measured Frequency Range
Min
Max
Error Range (PPM)
Min
Max
Program Calibration Register to:
Positive Calibration for slow clocks: Calibration will achieve ±2.17 PPM after calibration
0
512.0000
511.9989
0
2.17
000000
1
511.9989
511.9967
2.18
6.51
100001
2
511.9967
511.9944
6.52
10.85
100010
3
511.9944
511.9922
10.86
15.19
100011
4
511.9922
511.9900
15.20
19.53
100100
5
511.9900
511.9878
19.54
23.87
100101
6
511.9878
511.9856
23.88
28.21
100110
7
511.9856
511.9833
28.22
32.55
100111
8
511.9833
511.9811
32.56
36.89
101000
9
511.9811
511.9789
36.90
41.23
101001
10
511.9789
511.9767
41.24
45.57
101010
11
511.9767
511.9744
45.58
49.91
101011
12
511.9744
511.9722
49.92
54.25
101100
13
511.9722
511.9700
54.26
58.59
101101
14
511.9700
511.9678
58.60
62.93
101110
15
511.9678
511.9656
62.94
67.27
101111
16
511.9656
511.9633
67.28
71.61
110000
17
511.9633
511.9611
71.62
75.95
110001
18
511.9611
511.9589
75.96
80.29
110010
19
511.9589
511.9567
80.30
84.63
110011
20
511.9567
511.9544
84.64
88.97
110100
21
511.9544
511.9522
88.98
93.31
110101
22
511.9522
511.9500
93.32
97.65
110110
23
511.9500
511.9478
97.66
101.99
110111
24
511.9478
511.9456
102.00
106.33
111000
25
511.9456
511.9433
106.34
110.67
111001
26
511.9433
511.9411
110.68
115.01
111010
27
511.9411
511.9389
115.02
119.35
111011
28
511.9389
511.9367
119.36
123.69
111100
29
511.9367
511.9344
123.70
128.03
111101
30
511.9344
511.9322
128.04
132.37
111110
31
511.9322
511.9300
132.38
136.71
111111
Document Number: 001-86392 Rev. *D
Page 12 of 36
FM31L276/FM31L278
Table 3. Digital Calibration Adjustments (continued)
S.No.
Measured Frequency Range
Min
Max
Error Range (PPM)
Min
Max
Program Calibration Register to:
Negative Calibration for fast clocks: Calibration will achieve ±2.17 PPM after calibration
0
512.0000
512.0011
0
2.17
000000
1
512.0011
512.0033
2.18
6.51
000001
2
512.0033
512.0056
6.52
10.85
000010
3
512.0056
512.0078
10.86
15.19
000011
4
512.0078
512.0100
15.20
19.53
000100
5
512.0100
512.0122
19.54
23.87
000101
6
512.0122
512.0144
23.88
28.21
000110
7
512.0144
512.0167
28.22
32.55
000111
8
512.0167
512.0189
32.56
36.89
001000
9
512.0189
512.0211
36.90
41.23
001001
10
512.0211
512.0233
41.24
45.57
001010
11
512.0233
512.0256
45.58
49.91
001011
12
512.0256
512.0278
49.92
54.25
001100
13
512.0278
512.0300
54.26
58.59
001101
14
512.0300
512.0322
58.60
62.93
001110
15
512.0322
512.0344
62.94
67.27
001111
16
512.0344
512.0367
67.28
71.61
010000
17
512.0367
512.0389
71.62
75.95
010001
18
512.0389
512.0411
75.96
80.29
010010
19
512.0411
512.0433
80.30
84.63
010011
20
512.0433
512.0456
84.64
88.97
010100
21
512.0456
512.0478
88.98
93.31
010101
22
512.0478
512.0500
93.32
97.65
010110
23
512.0500
512.0522
97.66
101.99
010111
24
512.0522
512.0544
102.00
106.33
011000
25
512.0544
512.0567
106.34
110.67
011001
26
512.0567
512.0589
110.68
115.01
011010
27
512.0589
512.0611
115.02
119.35
011011
28
512.0611
512.0633
119.36
123.69
011100
29
512.0633
512.0656
123.70
128.03
011101
30
512.0656
512.0678
128.04
132.37
011110
31
512.0678
512.0700
132.38
136.71
011111
Document Number: 001-86392 Rev. *D
Page 13 of 36
FM31L276/FM31L278
Register Map
The RTC and processor companion functions are accessed via 25 special function registers, which are mapped to a separate I2C
device ID. The interface protocol is described on I2C Interface on page 21. The registers contain timekeeping data, control bits, and
information flags. A description of each register follows the summary table.
Table 4. Register Map Summary Table
Nonvolatile =
Address
Battery-backed =
Data
D7
D6
D5
D4
D3
D2
D1
Function
D0
Range
18h
Serial Number Byte 7
Serial Number 7
FFh
17h
Serial Number Byte 6
Serial Number 6
FFh
16h
Serial Number Byte 5
Serial Number 5
FFh
15h
Serial Number Byte 4
Serial Number 4
FFh
14h
Serial Number Byte 3
Serial Number 3
FFh
13h
Serial Number Byte 2
Serial Number 2
FFh
12h
Serial Number Byte 1
Serial Number 1
FFh
11h
Serial Number Byte 0
Serial Number 0
FFh
10h
Counter 2 MSB
Event Counter 2 MSB
FFh
0Fh
Counter 2 LSB
Event Counter 2 LSB
FFh
0Eh
Counter 1 MSB
Event Counter 1 MSB
FFh
Event Counter 1 LSB
FFh
Counter 1 LSB
0Dh
0Ch
–
–
–
–
RC
CC
C2P
C1P
Event Count Control
0Bh
SNL
–
FC
WP1
WP0
VBC
–
VTP
Companion Control
0Ah
WDE
–
–
WDT4
WDT3
WDT2
WDT1
WDT0
09h
WTR
POR
LB
–
WR3
WR2
WR1
WR0
10 years
08h
10
months
0
Watchdog Control
Watchdog Restart/Flags
years
Years
00–99
months
Month
01–12
Date
01–31
Day
01–07
Hours
00–23
07h
0
0
06h
0
0
05h
0
0
04h
0
0
03h
0
10 minutes
minutes
Minutes
00–59
02h
0
10 seconds
seconds
Seconds
00–59
10 date
0
0
OSCEN reserved
00h
reserved
CALS
CAL4
day
0
10 hours
01h
CF
date
hours
CAL3
reserved reserved reserved
CAL2
CAL1
CAL0
CAL Control
CAL
W
R
RTC Control
Note When the device is first powered up and programmed, all timekeeping registers must be written because the battery-backed
register values cannot be guaranteed. The table below shows the default values of the non-volatile registers. All other register values
should be treated as unknown.
Document Number: 001-86392 Rev. *D
Page 14 of 36
FM31L276/FM31L278
Table 5. Default Register Values
Address
Hex Value
Address
Hex Value
18h
0x00
0Ah
0x1F
17h
0x00
08h
0x00
16h
0x00
07h
0x01
15h
0x00
06h
0x01
14h
0x00
05h
0x01
13h
0x00
04h
0x00
12h
0x00
03h
0x01
11h
0x00
02h
0x00
0Bh
0x00
01h
0x80
Document Number: 001-86392 Rev. *D
Page 15 of 36
FM31L276/FM31L278
Table 6. Register Description
Address
Description
18h
Serial Number Byte 7
D7
D6
D5
D4
D3
D2
D1
D0
SN.63
SN.62
SN.61
SN.60
SN.59
SN.58
SN.57
SN.56
Upper byte of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
17h
16h
Serial Number Byte 6
D7
D6
D5
D4
D3
D2
D1
D0
SN.55
SN.54
SN.53
SN.52
SN.51
SN.50
SN.49
SN.48
Byte 6 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
Serial Number Byte 5
D7
D6
D5
D4
D3
D2
D1
D0
SN.47
SN.46
SN.45
SN.44
SN.43
SN.42
SN.41
SN.40
Byte 5 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
15h
Serial Number Byte 4
D7
D6
D5
D4
D3
D2
D1
D0
SN.39
SN.38
SN.37
SN.36
SN.35
SN.34
SN.33
SN.32
Byte 4 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
14h
Serial Number Byte 3
D7
D6
D5
D4
D3
D2
D1
D0
SN.31
SN.30
SN.29
SN.28
SN.27
SN.26
SN.25
SN.24
Byte 3 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
13h
Serial Number Byte 2
D7
D6
D5
D4
D3
D2
D1
D0
SN.23
SN.22
SN.21
SN.20
SN.19
SN.18
SN.17
SN.16
Byte 2 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
12h
Serial Number Byte 1
D7
D6
D5
D4
D3
D2
D1
D0
SN.15
SN.14
SN.13
SN.12
SN.11
SN.10
SN.9
SN.8
Byte 1 of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
11h
Serial Number Byte 0
D7
D6
D5
D4
D3
D2
D1
D0
SN.7
SN.6
SN.5
SN.4
SN.3
SN.2
SN.1
SN.0
LSB of the serial number. Read/write when SNL = ‘0’, read-only when SNL = ‘1’. Nonvolatile.
10h
Counter 2 MSB
D7
D6
D5
D4
D3
D2
D1
D0
C2.15
C2.14
C2.13
C2.12
C2.11
C2.10
C2.9
C2.8
Event Counter 2 MSB. Increments on overflows from Counter 2 LSB. Battery-backed, read/write.
Document Number: 001-86392 Rev. *D
Page 16 of 36
FM31L276/FM31L278
Table 6. Register Description (continued)
Address
Description
0Fh
Counter 2 LSB
D7
D6
D5
D4
D3
D2
D1
D0
C2.7
C2.6
C2.5
C2.4
C2.3
C2.2
C2.1
C2.0
Event Counter 2 LSB. Increments on programmed edge event on CNT2 input or overflows from Counter 1 MSB when
CC = ‘1’. Battery-backed, read/write.
0Eh
Counter 1 MSB
D7
D6
D5
D4
D3
D2
D1
D0
C1.15
C1.14
C1.13
C1.12
C1.11
C1.10
C1.9
C1.8
Event Counter 1MSB. Increments on overflows from Counter 1 LSB. Battery-backed, read/write.
0Dh
Counter 1 LSB
D7
D6
D5
D4
D3
D2
D1
D0
C1.7
C1.6
C1.5
C1.4
C1.3
C1.2
C1.1
C1.0
Event Counter 1 LSB. Increments on programmed edge event on CNT1 input. Battery-backed, read/write.
0Ch
Event Counter Control
D7
D6
D5
D4
D3
D2
D1
D0
–
–
–
–
RC
CC
C2P
C1P
RC
Read Counter. Setting this bit to ‘1’ takes a snapshot of the four counters bytes allowing the system to read the values
without missing count events. The RC bit will be automatically cleared.
CC
Counter Cascade. When CC = ‘0’, the event counters operate independently according to the edge programmed by
C1P and C2P respectively. When CC = ‘1’, the counters are cascaded to create one 32-bit counter. The registers of
Counter 2 represent the most significant 16-bits of the counter and CNT1 is the controlling input. Bit C2P is don't care
when CC = ‘1’. Battery-backed, read/write.
C2P
CNT2 detects falling edges when C2P = ‘0’, rising edges when C2P = ‘1’. C2P is “don't care” when CC = ‘1’. The
value of Event Counter 2 may inadvertently increment if C2P is changed. Battery-backed, read/write.
C1P
CNT1 detects falling edges when C1P = ‘0’, rising edges when C1P = ‘1’. The value of Event Counter 1 may
inadvertently increment if C1P is changed. Battery-backed, read/write.
0Bh
Companion Control
D7
D6
D5
D4
D3
D2
D1
D0
SNL
–
FC
WP1
WP0
VBC
–
VTP
SNL
Serial Number Lock: Setting to a ‘1’ makes registers 11h to 18h and SNL permanently read-only. SNL cannot be
cleared once set to ‘1’. Nonvolatile, read/write.
FC
Fast Charge: Setting FC to ‘1’ (and VBC = ‘1’) causes a ~1 mA trickle charge current to be supplied on VBAK. Clearing
VBC to ‘0’ disables the charge current. Nonvolatile, read/write.
WP(1:0)
Write Protect. These bits control the write protection of the memory array. Nonvolatile, read/write.
VBC
Write protect address
WP1
WP0
None
0
0
Bottom 1/4
0
1
Bottom 1/2
1
0
Full array
1
1
VBAK Charger Control. Setting VBC to ‘1’ (and FC = ‘0’) causes a 80 µA (1 mA if FC = ‘1’) trickle charge current to
be supplied on VBAK. Clearing VBC to ‘0’ disables the charge current. Nonvolatile, read/write.
Document Number: 001-86392 Rev. *D
Page 17 of 36
FM31L276/FM31L278
Table 6. Register Description (continued)
Address
VTP
Description
VTP Select. This bit control the reset trip point for the low VDD reset function. Nonvolatile, read/write.
Trip Voltage
VTP
2.60 V
0
2.90 V
1
0Ah
Watchdog Control
D7
D6
D5
D4
D3
D2
D1
D0
WDE
–
–
WDT4
WDT3
WDT2
WDT1
WDT0
WDE
Watchdog Enable. When WDE = ‘1’, a watchdog timer fault will cause the RST signal to go active. When WDE = ‘0’
the timer runs but has no effect on RST, however the WTR flag will be set when a fault occurs. Note as the timer is
free-running, users should restart the timer using WR(3:0) prior to setting WDE = ‘1’. This assures a full watchdog
timeout interval occurs. Nonvolatile, read/write.
WDT(4:0)
Watchdog Timeout. Indicates the minimum watchdog timeout interval with 100 ms resolution. New watchdog timeouts
are loaded when the timer is restarted by writing the 1010b pattern to WR(3:0). Nonvolatile, read/write.
Watchdog Timeout
WDT4
WDT3
WDT2
WDT1
WDT0
Invalid - default 100 ms
0
0
0
0
0
100 ms
0
0
0
0
1
200 ms
0
0
0
1
0
300 ms
0
0
0
1
1
2000 ms
1
0
1
0
0
2100 ms
1
0
1
0
1
2200 ms
1
0
1
1
0
2900 ms
1
1
1
0
1
3000 ms
1
1
1
1
0
Disable Counter
1
1
1
1
1
.
.
.
.
09h
Watchdog Restart and Flags
D7
D6
D5
D4
D3
D2
D1
D0
WTR
POR
LB
–
WR3
WR2
WR1
WR0
WTR
Watchdog Timer Reset Flag: When a watchdog timer fault occurs, the WTR bit will be set to ‘1’. It must be cleared
by the user. Note that both WTR and POR could be set if both reset sources have occurred since the flags were
cleared by the user. Battery-backed. Read/Write (internally set, user can clear bit).
POR
Power-on Reset Flag: When the RST pin is activated by VDD < VTP, the POR bit will be set to ‘1’. It must be cleared
by the user. Note that both WTR and POR could be set if both reset sources have occurred since the flags were
cleared by the user. Battery-backed. Read/Write (internally set, user can clear bit).
LB
Low Backup Flag: On power up, if the VBAK source is below the minimum voltage to operate the RTC and event
counters, this bit will be set to ‘1’. The user should clear it to ‘0’ when initializing the system. Battery-backed.
Read/Write (internally set, user can clear bit).
WR(3:0)
Watchdog Restart: Writing a pattern 1010b to WR(3:0) restarts the watchdog timer. The upper nibble contents do not
affect this operation. Writing any pattern other than 1010b to WR(3:0) has no effect on the timer. This allows users
to clear the WTR, POR, and LB flags without affecting the watchdog timer. Battery-backed, Write-only.
Document Number: 001-86392 Rev. *D
Page 18 of 36
FM31L276/FM31L278
Table 6. Register Description (continued)
Address
Description
08h
Timekeeping – Years
D7
D6
D5
D4
D3
D2
D1
D0
10 year.3
10 year.2
10 year.1
10 year.0
Year.3
Year.2
Year.1
Year.0
Contains the lower two BCD digits of the year. Lower nibble contains the value for years; upper nibble contains the
value for 10s of years. Each nibble operates from 0 to 9. The range for the register is 0–99. Battery-backed, read/write.
07h
Timekeeping – Months
D7
D6
D5
D4
D3
D2
D1
D0
0
0
0
10 Month
Month.3
Month.2
Month.1
Month.0
Contains the BCD digits for the month. Lower nibble 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. Battery-backed,
read/write.
06h
Timekeeping – Date of the month
D7
D6
D5
D4
D3
D2
D1
D0
0
0
10 date.1
10 date.0
Date.3
Date.2
Date.1
Date.0
Contains the BCD digits for the date of the month. Lower nibble contains the lower digit and operates from 0 to 9;
upper nibble contains the upper digit and operates from 0 to 3. The range for the register is 1–31. Battery-backed,
read/write.
05h
Timekeeping – Day of the week
D7
D6
D5
D4
D3
D2
D1
D0
0
0
0
0
0
Day.2
Day.1
Day.0
Lower nibble 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, as the day is not integrated with the date.
Battery-backed, read/write.
04h
Timekeeping – Hours
D7
D6
D5
D4
D3
D2
D1
D0
0
0
10 hours.1
10 hours.0
Hours.3
Hours.2
Hours.1
Hours.0
Contains the BCD value of hours in 24-hour format. Lower nibble 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.
Battery-backed, read/write.
03h
Timekeeping – Minutes
D7
D6
D5
D4
D3
D2
D1
D0
0
10 min.2
10 min.1
10 min.0
Min.3
Min.2
Min.1
Min.0
Contains the BCD value of minutes. Lower nibble contains the lower digit and operates from 0 to 9; upper nibble
contains the upper minutes digit and operates from 0 to 5. The range for the register is 0–59. Battery-backed,
read/write.
02h
Timekeeping – Seconds
D7
D6
D5
D4
D3
D2
D1
D0
0
10 sec.2
10 sec.1
10 sec.0
Seconds.3
Seconds.2
Seconds.1
Seconds.0
Contains the BCD value of seconds. Lower nibble contains the lower digit and operates from 0 to 9; upper nibble
contains the upper digit and operates from 0 to 5. The range for the register is 0-59. Battery-backed, read/write.
01h
CAL/Control
D7
D6
D5
D4
D3
D2
D1
D0
OSCEN
Reserved
CALS
CAL.4
CAL.3
CAL.2
CAL.1
CAL.0
Document Number: 001-86392 Rev. *D
Page 19 of 36
FM31L276/FM31L278
Table 6. Register Description (continued)
Address
Description
OSCEN
Oscillator Enable. When set to ‘1’, the oscillator is halted. When set to ‘0’, the oscillator runs after tOSC time. Disabling
the oscillator can save battery power during storage. On a power-up without battery, this bit is set to ‘1’.
Battery-backed, read/write.
Reserved
Reserved bits. Do not use. Should remain set to ‘0’.
CALS
Calibration Sign: Determines if the calibration adjustment is applied as an addition to or as a subtraction from the
time-base. This bit can be written only when CAL = ‘1’. Nonvolatile, read/write.
CAL(4:0)
Calibration Setting: These five bits control the calibration of the clock. These bits can be written only when CAL = ‘1’.
Nonvolatile, read/write.
00h
RTC Control
D7
D6
D5
D4
D3
D2
D1
D0
Reserved
CF
Reserved
Reserved
Reserved
CAL
W
R
CF
Century Overflow Flag. This bit is set to a ‘1’ when the values in the years register overflows from 99 to 00. This
indicates a new century, such as going from 1999 to 2000 or 2099 to 2100. The user should record the new century
information as needed. This bit is cleared to ‘0’ when the Flag register is read. It is read-only for the user.
Battery-backed.
CAL
Calibration Setting. When set to ‘1’, the clock enters calibration mode. When CAL is set to ‘0’, the clock operates
normally, and the CAL/PFO pin is controlled by the power fail comparator. Battery-backed, read/write.
W
Write Time. Setting the W bit to ‘1’ freezes the clock. The user can then write the timekeeping registers with updated
values. Resetting the W bit to ‘0’ causes the contents of the time registers to be transferred to the timekeeping counters
and restarts the clock. Battery-backed, read/write.
R
Read Time. Setting the R bit to ‘1’ copies a static image of the timekeeping core and place it into the user registers.
The user can then read them without concerns over changing values causing system errors. The R bit going from ‘0’
to ‘1’ causes the timekeeping capture, so the bit must be returned to ‘0’ prior to reading again. Battery-backed,
read/write.
Reserved
Reserved bits. Do not use. Should remain set to ‘0’.
Document Number: 001-86392 Rev. *D
Page 20 of 36
FM31L276/FM31L278
I2C Interface
STOP Condition (P)
The FM31L276/FM31L278 employs an industry standard I2C
bus that is familiar to many users. This product is unique since it
incorporates two logical devices in one chip. Each logical device
can be accessed individually. Although monolithic, it appears to
the system software to be two separate products. One is a
memory device. It has a Slave Address (Slave ID = 1010b) that
operates the same as a stand-alone memory device. The second
device is a real-time clock and processor companion which have
a unique Slave Address (Slave ID = 1101b).
A STOP condition is indicated when the bus master drives SDA
from LOW to HIGH while the SCL signal is HIGH. All operations
using the FM31L276/FM31L278 should end with a STOP
condition. If an operation is in progress when a STOP is
asserted, the operation will be aborted. The master must have
control of SDA in order to assert a STOP condition.
By convention, any device that is sending data onto the bus is
the transmitter while the target device for this data is the receiver.
The device that is controlling the bus is the master. The master
is responsible for generating the clock signal for all operations.
Any device on the bus that is being controlled is a slave. The
FM31L276/FM31L278 is always a slave device.
The bus protocol is controlled by transition states in the SDA and
SCL signals. There are four conditions including START, STOP,
data bit, or acknowledge. Figure 12 and Figure 13 illustrates the
signal conditions that specify the four states. Detailed timing
diagrams are shown in the electrical specifications section.
START Condition (S)
A START condition is indicated when the bus master drives SDA
from HIGH to LOW while the SCL signal is HIGH. All commands
should be preceded by a START condition. An operation in
progress can be aborted by asserting a START condition at any
time. Aborting an operation using the START condition will ready
the FM31L276/FM31L278 for a new operation.
If during operation the power supply drops below the specified
VTP minimum, any I2C transaction in progress will be aborted
and the system should issue a START condition prior to
performing another operation.
Figure 12. START and STOP Conditions
full pagewidth
SDA
SDA
SCL
SCL
S
P
STOP Condition
START Condition
Data/Address Transfer
All data transfers (including addresses) take place while the SCL
signal is HIGH. Except under the three conditions described
above, the SDA signal should not change while SCL is HIGH.
Figure 13. Data Transfer on the I2C Bus
handbook, full pagewidth
P
SDA
Acknowledgement
signal from slave
MSB
SCL
S
1
2
7
8
9
ACK
START
condition
Document Number: 001-86392 Rev. *D
Byte complete
1
Acknowledgement
signal from receiver
2
3
4-8
9
ACK
S
S
or
P
STOP or
START
condition
Page 21 of 36
FM31L276/FM31L278
Acknowledge/No-acknowledge
The acknowledge takes place after the 8th data bit has been
transferred in any transaction. During this state the transmitter
should release the SDA bus to allow the receiver to drive it. The
receiver drives the SDA signal LOW to acknowledge receipt of
the byte. If the receiver does not drive SDA LOW, the condition
is a no-acknowledge and the operation is aborted.
The receiver would fail to acknowledge for two distinct reasons.
First is that a byte transfer fails. In this case, the no-acknowledge
ceases the current operation so that the device can be
addressed again. This allows the last byte to be recovered in the
event of a communication error.
Second and most common, the receiver does not acknowledge
to deliberately end an operation. For example, during a read
operation, the FM31L276/FM31L278 will continue to place data
onto the bus as long as the receiver sends acknowledges (and
clocks). When a read operation is complete and no more data is
needed, the receiver must not acknowledge the last byte. If the
receiver acknowledges the last byte, this will cause the
FM31L276/FM31L278 to attempt to drive the bus on the next
clock while the master is sending a new command such as
STOP.
Figure 14. Acknowledge on the I2C Bus
handbook, full pagewidth
DATA OUTPUT
BY MASTER
No Acknowledge
DATA OUTPUT
BY SLAVE
Acknowledge
SCL FROM
MASTER
1
2
8
9
S
Clock pulse for
acknowledgement
START
Condition
Slave Address
The first byte that the FM31L276/FM31L278 expects after a
START condition is the slave address. As shown in Figure 15
and Figure 16, the slave address contains the device type or
slave ID, the device select address bits, and a bit that specifies
if the transaction is a read or a write.
The FM31L276/FM31L278 has two Slave Addresses (Slave IDs)
associated with two logical devices. Bits 7–4 are the device type
(slave ID) and should be set to 1010b for the memory device.
The other logical device within the FM31L276/FM31L278 is the
real-time clock and companion. Bits 7–4 are the device type
(slave ID) and should be set to 1101b for the RTC and
companion. A bus transaction with this slave address will not
affect the memory in any way. The figures below illustrate the two
Slave Addresses.
Bits 2-1 are the device select address bits. They must match the
corresponding value on the external address pins to select the
device. Up to four FM31L276/FM31L278 devices can reside on
the same I2C bus by assigning a different address to each. Bit 0
Document Number: 001-86392 Rev. *D
is the read/write bit (R/W). R/W = ‘1’ indicates a read operation
and R/W = ‘0’ indicates a write operation.
Figure 15. Memory Slave Device Address
MSB
handbook, halfpage
1
LSB
0
1
0
X
A1
A0 R/W
Device
Select
Slave ID
Figure 16. Companion Slave Device Address
MSB
handbook, halfpage
1
LSB
1
0
Slave ID
1
X
A1
A0 R/W
Device
Select
Page 22 of 36
FM31L276/FM31L278
Addressing Overview - Memory
Addressing Overview - RTC & Companion
After the FM31L276/FM31L278 (as receiver) acknowledges the
slave address, the master can place the memory address on the
bus for a write operation. The address requires two bytes. The
complete 15-bit address is latched internally. Each access
causes the latched address value to be incremented automatically. The current address is the value that is held in the latch;
either a newly written value or the address following the last
access. The current address will be held for as long
as VDD > VTP or until a new value is written. Reads always use
the current address. A random read address can be loaded by
beginning a write operation as explained below.
The RTC and Processor Companion operate in a similar manner
to the memory, except that it uses only one byte of address.
Addresses 00h to 18h correspond to special function registers.
Attempting to load addresses above 18h is an illegal condition;
the FM31L276/FM31L278 will return a NACK and abort the I2C
transaction.
After transmission of each data byte, just prior to the
acknowledge, the FM31L276/FM31L278 increments the internal
address latch. This allows the next sequential byte to be
accessed with no additional addressing. After the last address
(7FFFh) is reached, the address latch will roll over to 0000h.
There is no limit to the number of bytes that can be accessed
with a single read or write operation.
Document Number: 001-86392 Rev. *D
Data Transfer
After the address bytes have been transmitted, data transfer
between the bus master and the FM31L276/FM31L278 can
begin. For a read operation the FM31L276/FM31L278 will place
8 data bits on the bus then wait for an acknowledge from the
master. If the acknowledge occurs, the FM31L276/FM31L278
will transfer the next sequential byte. If the acknowledge is not
sent, the FM31L276/FM31L278 will end the read operation. For
a write operation, the FM31L276/FM31L278 will accept 8 data
bits from the master then send an acknowledge. All data transfer
occurs MSB (most significant bit) first.
Page 23 of 36
FM31L276/FM31L278
Memory Operation
bytes may be written. If the end of the address range is reached
internally, the address counter will wrap from 7FFFh to 0000h.
The FM31L276/FM31L278 is designed to operate in a manner
very similar to other I2C interface memory products. The major
differences result from the higher performance write capability of
F-RAM technology. These improvements result in some
differences between the FM31L276/FM31L278 and a similar
configuration EEPROM during writes. The complete operation
for both writes and reads is explained below.
Unlike other nonvolatile memory technologies, there is no
effective write delay with F-RAM. Since the read and write
access times of the underlying memory are the same, the user
experiences no delay through the bus. The entire memory cycle
occurs in less time than a single bus clock. Therefore, any
operation including read or write can occur immediately following
a write. Acknowledge polling, a technique used with EEPROMs
to determine if a write is complete is unnecessary and will always
return a ready condition.
The memory address for FM31L276 range from 0x0000 to
0x1FFFF, and for FM31L278, they range from 0x0000 to 0x7FFF.
Memory functionality is described with respect to FM31L278 in
the following sections.
Internally, an actual memory write occurs after the 8th data bit is
transferred. It will be complete before the acknowledge is sent.
Therefore, if the user desires to abort a write without altering the
memory contents, this should be done using START or STOP
condition prior to the 8th data bit. The FM31L276/FM31L278
uses no page buffering.
Memory Write Operation
All writes begin with a slave address, then a memory address.
The bus master indicates a write operation by setting the LSB of
the slave address (R/W bit) to a ‘0’. After addressing, the bus
master sends each byte of data to the memory and the memory
generates an acknowledge condition. Any number of sequential
Figure 17 and Figure 18 below illustrate a single-byte and
multiple-byte write cycles.
Figure 17. Single-Byte Write
By Master
Start
S
Stop
Address & Data
Slave Address
0 A
Address MSB
By F-RAM
A
Address LSB
A
Data Byte
A
P
Acknowledge
Figure 18. Multi-Byte Write
Start
Stop
Address & Data
By Master
S
Slave Address
0 A
By F-RAM
Document Number: 001-86392 Rev. *D
Address MSB
A
Address LSB
A
Data Byte
A
Data Byte
A
P
Acknowledge
Page 24 of 36
FM31L276/FM31L278
Memory Read Operation
There are two basic types of read operations. They are current
address read and selective address read. In a current address
read, the FM31L276/FM31L278 uses the internal address latch
to supply the address. In a selective read, the user performs a
procedure to set the address to a specific value.
Current Address & Sequential Read
As mentioned above the FM31L276/FM31L278 uses an internal
latch to supply the address for a read operation. A current
address read uses the existing value in the address latch as a
starting place for the read operation. The system reads from the
address immediately following that of the last operation.
To perform a current address read, the bus master supplies a
slave address with the LSB set to a ‘1’. This indicates that a read
operation is requested. After receiving the complete slave
address, the FM31L276/FM31L278 will begin shifting out data
from the current address on the next clock. The current address
is the value held in the internal address latch.
Beginning with the current address, the bus master can read any
number of bytes. Thus, a sequential read is simply a current
address read with multiple byte transfers. After each byte the
internal address counter will be incremented.
Note Each time the bus master acknowledges a byte, this
indicates that the FM31L276/FM31L278 should read out the next
sequential byte.
There are four ways to properly terminate a read operation.
Failing to properly terminate the read will most likely create a bus
contention as the FM31L276/FM31L278 attempts to read out
additional data onto the bus. The four valid methods are:
1. The bus master issues a no-acknowledge in the 9th clock
cycle and a STOP in the 10th clock cycle. This is illustrated in
the diagrams below. This is preferred.
2. The bus master issues a no-acknowledge in the 9th clock
cycle and a START in the 10th.
3. The bus master issues a STOP in the 9th clock cycle.
4. The bus master issues a START in the 9th clock cycle.
If the internal address reaches 7FFFh, it will wrap around to
0000h on the next read cycle. Figure 19 and Figure 20 below
show the proper operation for current address reads.
Figure 19. Current Address Read
By Master
Start
No
Acknowledge
Address
Stop
S
Slave Address
By F-RAM
1 A
Acknowledge
Data Byte
1
P
Data
Figure 20. Sequential Read
By Master
Start
No
Acknowledge
Acknowledge
Address
Stop
S
Slave Address
By F-RAM
Document Number: 001-86392 Rev. *D
1 A
Acknowledge
Data Byte
A
Data Byte
1 P
Data
Page 25 of 36
FM31L276/FM31L278
sends the address bytes that are loaded into the internal address
latch. After the FM31L276/FM31L278 acknowledges the
address, the bus master issues a START condition. This simultaneously aborts the write operation and allows the read
command to be issued with the slave address LSB set to a '1'.
The operation is now a current address read.
Selective (Random) Read
There is a simple technique that allows a user to select a random
address location as the starting point for a read operation. This
involves using the first three bytes of a write operation to set the
internal address followed by subsequent read operations.
To perform a selective read, the bus master sends out the slave
address with the LSB (R/W) set to ‘0’. This specifies a write
operation. According to the write protocol, the bus master then
Figure 21. Selective (Random) Read
Start
Address
By Master
Start
No
Acknowledge
Address
Stop
S
Slave Address
0 A
Address MSB
A
Address LSB
By F-RAM
A
S
Slave Address
1 A
Data Byte
Data
Acknowledge
RTC/Companion Write Operation
1 P
Note Although not required, it is recommended that A5-A7 in the
register address byte are zeros in order to preserve compatibility
with future devices.
All RTC and Companion writes operate in a similar manner to
memory writes. The distinction is that a different device ID is
used and only one byte address is needed instead of two byte
address. Figure 22 illustrates a single byte write to this device.
Figure 22. Single Byte Write
By Master
Address & Data
Start
S
Slave Address
0 A 0 0 0
Address
By F-RAM
Stop
A
Data Byte
A
P
Acknowledge
RTC/Companion Read Operation
As with writes, a read operation begins with the Slave Address.
To perform a register read, the bus master supplies a Slave
Address with the LSB set to ‘1’. This indicates that a read
operation is requested. After receiving the complete Slave
Address, the FM31L276/FM31L278 will begin shifting data out
from the current register address on the next clock.
Auto-increment operates for the special function registers as
with the memory address. A current address read for the
registers look exactly like the memory except that the device ID
is different.
The FM31L276/FM31L278 contains two separate address
registers, one for the memory address and the other for the
register address. This allows the contents of one address
register to be modified without affecting the current address of
the other register. For example, this would allow an interrupted
read to the memory while still providing fast access to an RTC
register. A subsequent memory read will then continue from the
memory address where it previously left off, without requiring the
load of a new memory address. However, a write sequence
always requires an address to be supplied.
Addressing FRAM Array in the FM31L276/FM31L278
Family
The FM31L276/FM31L278 family includes 64-Kbit and 256-Kbit
memory densities. The following 2-byte address field is shown
for each density.
1st Address Byte
Part Number
2nd Address Byte
FM31L276
X
X
X
A12
A11
A10
A9
A8
A7
A6
A5
A4
A3
A2
A1
A0
FM31L278
X
A14
A13
A12
A11
A10
A9
A8
A7
A6
A5
A4
A3
A2
A1
A0
Document Number: 001-86392 Rev. *D
Page 26 of 36
FM31L276/FM31L278
Maximum Ratings
Exceeding maximum ratings may shorten the useful life of the
device. These user guidelines are not tested.
Storage temperature ................................ –65 °C to +125 °C
Ambient Temperature
with power applied ................................... –55 C to +125 C
Supply voltage on VDD relative to VSS .........–1.0 V to +5.0 V
Surface mount lead soldering temperature
(3 seconds) .............................................................. +260 °C
DC output current
(1 output at a time, 1s duration) .................................. 15 mA
Electrostatic Discharge Voltage
Human Body Model (AEC-Q100-002 Rev. D) ....................... 2 kV
Charged Device Model (AEC-Q100-011 Rev. B) .............. 1.25 kV
Machine Model (AEC-Q100-003 Rev. E) ............................. 100 V
Input voltage ........... –1.0 V to +5.0 V and VIN < VDD + 1.0 V
Latch-up current .................................................. > ±100 mA
Backup supply voltage .................................–1.0 V to +4.5 V
Note PFI input voltage must not exceed 4.5 V. The
“VIN < VDD + 1.0 V” restriction does not apply to the SCL and
SDA inputs which do not employ a diode to VDD.
DC voltage applied to outputs
in High-Z state .................................... –0.5 V to VDD + 0.5 V
Transient voltage (< 20 ns)
on any pin to ground potential ............ –2.0 V to VDD + 2.0 V
Package power dissipation capability
(TA = 25 °C) ................................................................. 1.0 W
Operating Range
Range
Industrial
Ambient Temperature (TA)
–40 C to +85 C
VDD
2.7 V to 3.6 V
DC Electrical Characteristics
Over the Operating Range
Parameter
VDD
[2]
IDD
Description
Average VDD current
VDD standby current
VBAK[3]
RTC backup voltage
IBAKTC[4]
Typ [1]
Max
Unit
2.7
–
3.6
V
fSCL = 100 kHz
–
–
500
A
fSCL = 400 kHz
–
–
900
A
fSCL = 1 MHz
–
–
1500
A
–
–
120
A
TA = +25 C to +85 C
1.55
–
3.75
V
TA = –40 C to +25 C
1.90
–
3.75
V
TA = +25 C,
VBAK = 3.0 V
–
–
1.4
A
TA = +85 C,
VBAK = 3.0 V
–
–
2.1
A
TA = +25 C,
VBAK = 2.0 V
–
–
1.15
A
TA = +85 C,
VBAK = 2.0 V
–
–
1.75
A
Fast Charge Off
(FC = ‘0’)
50
–
120
A
Fast Charge On
(FC = ‘1’)
200
–
2500
A
Power supply
ISB
IBAK
Min
Test Conditions
RTC backup current
Trickle charge current with
VBAK = 0 V
SCL toggling between
VDD – 0.3 V and VSS,
other inputs
VSS or VDD – 0.3 V.
SCL = SDA = VDD.
All other inputs
VSS or VDD.
Stop
command
issued.
VBAK = 3.0 V,
VDD < 2.4 V,
oscillator running,
CNT1, CNT2 at VBAK.
Notes
1. Typical values are at 25 °C, VDD = VDD(typ). Not 100% tested.
2. Full complete operation. Supervisory circuits, RTC, etc operate to lower voltages as specified.
3. The VBAK trickle charger automatically regulates the maximum voltage on this pin for capacitor backup applications.
4. VBAK will source current when trickle charge is enabled (VBC bit = ‘1’), VDD > VBAK, and VBAK < VBAK max.
Document Number: 001-86392 Rev. *D
Page 27 of 36
FM31L276/FM31L278
DC Electrical Characteristics (continued)
Over the Operating Range
Parameter
Description
Test Conditions
Min
Typ [1]
Max
Unit
VTP1
VDD trip point voltage, VTP = 0
RST is asserted active
when VDD < VTP.
2.55
2.60
2.70
V
VTP2
VDD trip point voltage, VTP = 1
RST is asserted active
when VDD < VTP.
2.85
2.90
3.00
V
VRST[5]
VDD for valid RST
IOL = 80 A at VOL
VBAK > VBAK min
0
–
–
V
VBAK < VBAK min
1.6
–
–
V
ILI
Input leakage current
VSS < VIN < VDD. Does not apply to A0, A1, X1,
PFI, RST, or X2
–
–
±1
A
ILO
Output leakage current
VSS < VOUT < VDD. Does not apply to RST, X1,
or X2
–
–
±1
A
VIL[6]
Input LOW voltage
All inputs except as
listed below
– 0.3
–
0.3 × VDD
V
CNT1, CNT2
battery-backed
(VDD < 2.5 V)
– 0.3
–
0.5
V
CNT1, CNT2
(VDD > 2.5 V)
– 0.3
–
0.8
V
All inputs except as
listed below
0.7 × VDD
–
VDD + 0.3
V
CNT1, CNT2
battery-backed
(VDD < 2.5 V)
VBAK – 0.5
–
VBAK + 0.3
V
0.7 × VDD
–
VDD + 0.3
V
–
–
3.75
V
2.4
–
–
V
VIH
Input HIGH voltage
CNT1, CNT2
(VDD > 2.5 V)
PFI (comparator input)
VOH
Output HIGH voltage
IOH = –2 mA
VOL
Output LOW voltage
IOL = 3 mA
RRST
Pull-up resistance for RST
inactive
Rin
Input resistance (A1-A0)
VPFI
Power fail input reference voltage
VHYS
Power fail input (PFI) hysteresis
(rising)
For VIN = VIL(Max)
For VIN = VIH(Min)
–
–
0.4
V
50
–
400
k
20
–
–
k
1
–
–
M
1.175
1.20
1.225
V
–
–
100
mV
Notes
5. The minimum VDD to guarantee the level of RST remains a valid VOL level.
6. Includes RST input detection of external reset condition to trigger driving of RST signal by FM31L276/FM31L278.
Document Number: 001-86392 Rev. *D
Page 28 of 36
FM31L276/FM31L278
Data Retention and Endurance
Parameter
TDR
NVC
Description
Test condition
Data retention
Endurance
Min
Max
Unit
TA = 85 C
10
–
Years
TA = 75 C
38
–
TA = 65 C
151
–
14
–
Cycles
Typ
Max
Unit
–
8
pF
12
–
pF
Test Conditions
14-pin SOIC
Unit
Test conditions follow standard test methods and
procedures for measuring thermal impedance, per EIA /
JESD51.
80
C/W
29
C/W
Over operating temperature
10
Capacitance
Parameter [7]
Description
Test Conditions
CIO
Input/Output pin capacitance
CXTL[8]
X1, X2 crystal pin capacitance
TA = 25 C, f = 1 MHz, VDD = VDD(typ)
Thermal Resistance
Parameter [7]
Description
JA
Thermal resistance
(junction to ambient)
JC
Thermal resistance
(junction to case)
AC Test Loads and Waveforms
Figure 23. AC Test Loads and Waveforms
3.6V
1.66 k
OUTPUT
100 pF
AC Test Conditions
Input pulse levels .................................10% and 90% of VDD
Input rise and fall times .................................................10 ns
Input and output timing reference levels ................0.5 × VDD
Output load capacitance ............................................ 100 pF
Notes
7. This parameter is characterized and not 100% tested.
8. The crystal attached to the X1/X2 pins must be rated as 6 pF/12.5 pF.
Document Number: 001-86392 Rev. *D
Page 29 of 36
FM31L276/FM31L278
Supervisor Timing
Over the Operating Range
Description
Min
Max
Units
RST active (LOW) after VDD > VTP
100
200
ms
RST response time to VDD < VTP (noise filter)
10
25
s
–
s/V
Parameter
tRPU
tRNR
tVR
[9]
[9, 10]
VDD power-up ramp rate
50
tVF[9, 10]
tWDP[11]
tWDOG[11]
VDD power-down ramp rate
100
–
s/V
Pulse width of RST for watchdog reset
100
200
ms
Timeout of watchdog
tDOG
2 × tDOG
ms
fCNT
Frequency of event counters
0
10
MHz
tOSC
RTC Oscillator time to start
–
2
s
Figure 24. RST Timing
t VF
V DD
V TP
V RST
t VR
tRNR
tRPU
RST
Notes
9. This parameter is characterized and not 100% tested.
10. Slope measured at any point on VDD waveform.
11. tDOG is the programmed time in register in register 0Ah, VDD > VTP, and tRPU satisfied.
Document Number: 001-86392 Rev. *D
Page 30 of 36
FM31L276/FM31L278
AC Switching Characteristics
Over the Operating Range
Parameter[12]
Cypress
Parameter
Description
Alt.
Parameter
Min
Max
Min
Max
Min
Max
Unit
0
100
0
400
0
1000
kHz
fSCL
SCL clock frequency
tSU; STA
Start condition setup for repeated Start
4.7
–
0.6
–
0.25
–
s
tHD;STA
Start condition hold time
4.0
–
0.6
–
0.25
–
s
tLOW
Clock LOW period
4.7
–
1.3
–
0.6
–
s
tHIGH
Clock HIGH period
4.0
–
0.6
–
0.4
–
s
tSU;DAT
tSU;DATA
Data in setup
250
–
100
–
100
–
ns
tHD;DAT
tHD;DATA
Data in hold
0
–
0
–
0
–
ns
Data output hold (from SCL @ VIL)
0
–
0
–
0
–
ns
–
300
–
300
ns
300
tDH
[13]
tr
Input rise time
–
1000
tF[13]
tf
Input fall time
–
300
–
STOP condition setup
4
–
0.6
SCL LOW to SDA Data Out Valid
–
3
–
tBUF
Bus free before new transmission
4.7
–
tSP
Noise suppression time constant on SCL, SDA
–
50
tR
tSU;STO
tAA
tVD;DATA
–
100
ns
0.25
–
s
0.9
–
0.55
s
1.3
–
0.5
–
s
–
50
–
50
ns
Figure 25. Read Bus Timing Diagram
tR
`
tF
tHIGH
tSP
tLOW
tSP
SCL
tSU:SDA
1/fSCL
tBUF
tHD:DAT
tSU:DAT
SDA
tDH
tAA
Stop Start
Start
Acknowledge
Figure 26. Write Bus Timing Diagram
tHD:DAT
SCL
tHD:STA
tSU:STO
tSU:DAT
tAA
SDA
Start
Stop Start
Acknowledge
Notes
12. Test conditions assume a signal transition time of 10 ns or less, timing reference levels of 0.5 × VDD, input pulse levels of 10% to 90% of VDD, and output loading of
the specified IOL/IOH and 100 pF load capacitance shown in Figure 23 on page 29.
13. This parameter is characterized and not 100% tested.
Document Number: 001-86392 Rev. *D
Page 31 of 36
FM31L276/FM31L278
Ordering Information
Ordering Code
Package
Diagram
Package Type
FM31L276-G
51-85067
14-pin SOIC
FM31L276-GTR
51-85067
14-pin SOIC
FM31L278-G
51-85067
14-pin SOIC
FM31L278-GTR
51-85067
14-pin SOIC
Operating
Range
Industrial
All these parts are Pb-free. Contact your local Cypress sales representative for availability of these parts.
Ordering Code Definitions
FM 31
L XXX - G
X
Option: X = blank or TR
blank = Standard; TR = Tape and Reel
Package Type:
G = 14-pin SOIC
Density: XXX = 276 or 278
276 = 64-Kbit; 278 = 256-Kbit
L: 2.7 V to 3.6 V
I2C Processor Companion
Cypress
Document Number: 001-86392 Rev. *D
Page 32 of 36
FM31L276/FM31L278
Package Diagram
Figure 27. 14-pin SOIC (150 Mils) Package Outline, 51-85067
51-85067 *E
Document Number: 001-86392 Rev. *D
Page 33 of 36
FM31L276/FM31L278
Acronyms
Acronym
Document Conventions
Description
Units of Measure
EEPROM
Electrically Erasable Programmable Read-Only
Memory
°C
degree Celsius
EIA
Electronic Industries Alliance
Hz
hertz
F-RAM
Ferroelectric Random Access Memory
kHz
kilohertz
I2C
Inter-Integrated Circuit
k
kilohm
I/O
Input/Output
Mbit
megabit
JEDEC
Joint Electron Devices Engineering Council
MHz
megahertz
JESD
JEDEC Standards
A
microampere
LSB
Least Significant Bit
F
microfarad
MSB
Most Significant Bit
s
microsecond
NMI
Non Maskable interrupt
mA
milliampere
RoHS
Restriction of Hazardous Substances
ms
millisecond
SOIC
Small Outline Integrated Circuit
ns
nanosecond
ohm
%
percent
pF
picofarad
V
volt
W
watt
Document Number: 001-86392 Rev. *D
Symbol
Unit of Measure
Page 34 of 36
FM31L276/FM31L278
Document History Page
Document Title: FM31L276/FM31L278, 64-Kbit/256-Kbit Integrated Processor Companion with F-RAM
Document Number: 001-86392
Rev.
ECN No.
Orig. of
Change
Submission
Date
**
3916894
GVCH
02/28/2013
New spec.
*A
4333096
GVCH
05/05/2014
Updated Real-time Clock Operation:
Updated Crystal Oscillator:
Updated description (Added use of 6 pF crystal).
Updated Maximum Ratings:
Removed “Package Moisture Sensitivity Level” and its corresponding details.
Added “Maximum junction temperature” and its corresponding details.
Added “DC voltage applied to outputs in High-Z state” and its corresponding
details.
Added “Transient voltage (< 20 ns) on any pin to ground potential” and its
corresponding details.
Added “Package power dissipation capability (TA = 25 °C)” and its
corresponding details.
Added “DC output current (1 output at a time, 1s duration)” and its
corresponding details.
Added “Latch-up current” and its corresponding details.
Updated Data Retention and Endurance:
Updated details corresponding to TDR parameter (Removed details
corresponding to Test Condition “TA = 80 °C”; added details corresponding to
Test Condition “TA = 65 °C”).
Added NVC parameter and its details.
Updated Capacitance:
Changed typical value of CXTL parameter from 25 pF to 12 pF.
Added Thermal Resistance.
Updated Package Diagram:
Removed Package Marking Scheme.
Updated to Cypress template.
*B
4562106
GVCH
11/05/2014
Updated Functional Description:
Added “For a complete list of related documentation, click here.” at the end.
*C
6083685
GVCH
02/28/2018
Updated Register Map:
Updated Table 6:
Removed “VTP1” bit from bit 1 corresponding to address “0Bh”.
Replaced “VTP0” with “VTP” in bit 0 corresponding to address “0Bh”.
Updated details in “Description” column corresponding to OSCEN (to add tOSC
time in the 01h register).
Updated Maximum Ratings:
Removed “Maximum junction temperature” and its corresponding details.
Added “Ambient Temperature with power applied” and its corresponding
details.
Updated Supervisor Timing:
Added tOSC parameter and its corresponding details.
Updated Package Diagram:
spec 51-85067 – Changed revision from *D to *E.
Updated to new template.
*D
6388437
GVCH
11/19/2018
Updated Maximum Ratings:
Replaced “–55 °C to +125 °C” with “–65 °C to +125 °C” in ratings corresponding
to “Storage temperature”.
Updated to new template.
Document Number: 001-86392 Rev. *D
Description of Change
Page 35 of 36
FM31L276/FM31L278
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.
PSoC® Solutions
Products
Arm® Cortex® Microcontrollers
Automotive
cypress.com/arm
cypress.com/automotive
Clocks & Buffers
Interface
cypress.com/clocks
cypress.com/interface
Internet of Things
Memory
cypress.com/iot
cypress.com/memory
Microcontrollers
cypress.com/mcu
PSoC
cypress.com/psoc
Power Management ICs
Cypress Developer Community
Community | Projects | Video | Blogs | Training | Components
Technical Support
cypress.com/support
cypress.com/pmic
Touch Sensing
cypress.com/touch
USB Controllers
Wireless Connectivity
PSoC 1 | PSoC 3 | PSoC 4 | PSoC 5LP | PSoC 6 MCU
cypress.com/usb
cypress.com/wireless
© Cypress Semiconductor Corporation, 2013–2018. This document is the property of Cypress Semiconductor Corporation and its subsidiaries, including Spansion LLC (“Cypress”). This document,
including any software or firmware included or referenced in this document (“Software”), is owned by Cypress under the intellectual property laws and treaties of the United States and other countries
worldwide. Cypress reserves all rights under such laws and treaties and does not, except as specifically stated in this paragraph, grant any license under its patents, copyrights, trademarks, or other
intellectual property rights. If the Software is not accompanied by a license agreement and you do not otherwise have a written agreement with Cypress governing the use of the Software, then Cypress
hereby grants you a personal, non-exclusive, nontransferable license (without the right to sublicense) (1) under its copyright rights in the Software (a) for Software provided in source code form, to
modify and reproduce the Software solely for use with Cypress hardware products, only internally within your organization, and (b) to distribute the Software in binary code form externally to end users
(either directly or indirectly through resellers and distributors), solely for use on Cypress hardware product units, and (2) under those claims of Cypress’s patents that are infringed by the Software (as
provided by Cypress, unmodified) to make, use, distribute, and import the Software solely for use with Cypress hardware products. Any other use, reproduction, modification, translation, or compilation
of the Software is prohibited.
TO THE EXTENT PERMITTED BY APPLICABLE LAW, CYPRESS MAKES NO WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, WITH REGARD TO THIS DOCUMENT OR ANY SOFTWARE
OR ACCOMPANYING HARDWARE, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. No computing
device can be absolutely secure. Therefore, despite security measures implemented in Cypress hardware or software products, Cypress does not assume any liability arising out of any security breach,
such as unauthorized access to or use of a Cypress product. In addition, the products described in these materials may contain design defects or errors known as errata which may cause the product
to deviate from published specifications. To the extent permitted by applicable law, Cypress reserves the right to make changes to this document without further notice. Cypress does not assume any
liability arising out of the application or use of any product or circuit described in this document. Any information provided in this document, including any sample design information or programming
code, is provided only for reference purposes. It is the responsibility of the user of this document to properly design, program, and test the functionality and safety of any application made of this
information and any resulting product. Cypress products are not designed, intended, or authorized for use as critical components in systems designed or intended for the operation of weapons, weapons
systems, nuclear installations, life-support devices or systems, other medical devices or systems (including resuscitation equipment and surgical implants), pollution control or hazardous substances
management, or other uses where the failure of the device or system could cause personal injury, death, or property damage (“Unintended Uses”). A critical component is any component of a device
or system whose failure to perform can be reasonably expected to cause the failure of the device or system, or to affect its safety or effectiveness. Cypress is not liable, in whole or in part, and you
shall and hereby do release Cypress from any claim, damage, or other liability arising from or related to all Unintended Uses of Cypress products. You shall indemnify and hold Cypress harmless from
and against all claims, costs, damages, and other liabilities, including claims for personal injury or death, arising from or related to any Unintended Uses of Cypress products.
Cypress, the Cypress logo, Spansion, the Spansion logo, and combinations thereof, WICED, PSoC, CapSense, EZ-USB, F-RAM, and Traveo are trademarks or registered trademarks of Cypress in
the United States and other countries. For a more complete list of Cypress trademarks, visit cypress.com. Other names and brands may be claimed as property of their respective owners.
Document Number: 001-86392 Rev. *D
Revised November 19, 2018
Page 36 of 36
工商网监
湘ICP备2023018690号
