Document Number: 001-86211 Rev. *C Page 4 of 18
Device Operation
The FM1608B is a bytewide F-RAM memory logically organized
as 8,192 × 8 and accessed using an industry-standard parallel
interface. All data written to the part is immediately nonvolatile
with no delay. Functional operation of the F-RAM memory is the
same as SRAM type devices, except the FM1608B requires a
falling edge of CE to start each memory cycle. See the
Functional Truth Table on page 13 for a complete description of
read and write modes.
Memory Architecture
Users access 8,192 memory locations, each with 8 data bits
through a parallel interface. The complete 13-bit address
specifies each of the 8,192 bytes uniquely. The F-RAM array is
organized as 1024 rows of 8-bytes each. This row segmentation
has no effect on operation, however the user can group data into
blocks by its endurance characteristics as explained in the
Endurance section.
The cycle time is the same for read and write memory
operations. This simplifies memory controller logic and timing
circuits. Likewise the access time is the same for read and write
memory operations. When CE is deasserted HIGH, a pre-charge
operation begins, and is required of every memory cycle. Thus
unlike SRAM, the access and cycle times are not equal. Writes
occur immediately at the end of the access with no delay. Unlike
an EEPROM, it is not necessary to poll the device for a ready
condition since writes occur at bus speed.
It is the user’s responsibility to ensure that VDD remains within
datasheet tolerances to prevent incorrect operation. Also proper
voltage level and timing relationships between VDD and CE must
be maintained during power-up and power-down events. See
“Power Cycle Timing” on page 12.
Memory Operation
The FM1608B is designed to operate in a manner similar to other
bytewide memory products. For users familiar with BBSRAM, the
performance is comparable but the bytewide interface operates
in a slightly different manner as described below. For users
familiar with EEPROM, the differences result from the higher
write performance of F-RAM technology including NoDelay
writes and much higher write endurance.
Read Operation
A read operation begins on the falling edge of CE. At this time,
the address bits are latched and a memory cycle is initiated.
Once started, a full memory cycle must be completed internally
even if CE goes inactive. Data becomes available on the bus
after the access time is met.
After the address has been latched, the address value may be
changed upon satisfying the hold time parameter. Unlike an
SRAM, changing address values will have no effect on the
memory operation after the address is latched.
The FM1608B will drive the data bus when OE is asserted LOW
and the memory access time is met. If OE is asserted after the
memory access time is met, the data bus will be driven with valid
data. If OE is asserted before completing the memory access,
the data bus will not be driven until valid data is available. This
feature minimizes supply current in the system by eliminating
transients caused by invalid data being driven to the bus. When
OE is deasserted HIGH, the data bus will remain in a HI-Z state.
Write Operation
In the FM1608B, writes occur in the same interval as reads. The
FM1608B supports both CE and WE controlled write cycles. In
both cases, the address is latched on the falling edge of CE.
In a CE-controlled write, the WE signal is asserted before
beginning the memory cycle. That is, WE is LOW when the
device is activated with the chip enable. In this case, the device
begins the memory cycle as a write. The FM1608B will not drive
the data bus regardless of the state of OE.
In a WE-controlled write, the memory cycle begins on the falling
edge of CE. The WE signal falls after the falling edge of CE.
Therefore, the memory cycle begins as a read. The data bus will
be driven according to the state of OE until WE falls. The CE and
WE controlled write timing cases are shown in the page 12.
Write access to the array begins asynchronously after the
memory cycle is initiated. The write access terminates on the
rising edge of WE or CE, whichever comes first. A valid write
operation requires the user to meet the access time specification
before deasserting WE or CE. The data setup time indicates the
interval during which data cannot change before the end of the
write access.
Unlike other nonvolatile memory technologies, there is no write
delay with F-RAM. Because the read and write access times of
the underlying memory are the same, the user experiences no
delay through the bus. The entire memory operation occurs in a
single bus cycle. Therefore, any operation including read or write
can occur immediately following a write. Data polling, a
technique used with EEPROMs to determine if a write is
complete, is unnecessary.
Pre-charge Operation
The pre-charge operation is an internal condition in which the
memory state is prepared for a new access. All memory cycles
consist of a memory access and a pre-charge. Pre-charge is
user-initiated by driving the CE signal HIGH. It must remain
HIGH for at least the minimum pre-charge time, tPC.
The user determines the beginning of this operation since a
pre-charge will not begin until CE rises. However, the device has
a maximum CE LOW time specification that must be satisfied.
Endurance
Internally, a F-RAM operates with a read and restore
mechanism. Therefore, each read and write cycle involves a
change of state. The memory architecture is based on an array
of rows and columns. Each read or write access causes an
endurance cycle for an entire row. In the FM1608B, a row is 64
bits wide. Every 8-byte boundary marks the beginning of a new
row. Endurance can be optimized by ensuring frequently
accessed data is located in different rows. Regardless, F-RAM