STK14D88

To initiate the software STORE cycle, the following READ sequence must be performed:

1.Read Address 0x0E38, Valid READ

2.Read Address 0x31C7, Valid READ

3.Read Address 0x03E0, Valid READ

4.Read Address 0x3C1F, Valid READ

5.Read Address 0x303F, Valid READ

6.Read Address 0x0FC0, Initiate STORE Cycle

Once the sixth address in the sequence has been entered, the STORE cycle will commence and the chip will be disabled. It is important that READ cycles and not WRITE cycles be used in the sequence. After the tSTORE cycle time has been fulfilled, the SRAM will again be activated for READ and WRITE operation.

Software RECALL

Data can be transferred from the nonvolatile memory to the SRAM by a software address sequence. A software RECALL cycle is initiated with a sequence of READ operations in a manner similar to the software STORE initiation. To initiate the RECALL cycle, the following sequence of E controlled READ operations must be performed:

1.Read Address 0x0E38, Valid READ

2.Read Address 0x31C7, Valid READ

3.Read Address 0x03E0, Valid READ

4.Read Address 0x3C1F, Valid READ

5.Read Address 0x303F, Valid READ

6.Read Address 0x0C63, Initiate RECALL Cycle

Internally, RECALL is a two-step procedure. First, the SRAM data is cleared, and second, the nonvolatile information is trans- ferred into the SRAM cells. After the tRECALL cycle time, the SRAM will once again be ready for READ or WRITE operations. The RECALL operation in no way alters the data in the nonvol- atile storage elements.

Data Protection

The STK14D88 protects data from corruption during low-voltage conditions by inhibiting all externally initiated STORE and WRITE operations. The low-voltage condition is detected when

VCC<VSWITCH.

If the STK14D88 is in a WRITE mode (both E and W low) at power-up, after a RECALL, or after a STORE, the WRITE will be inhibited until a negative transition on E or W is detected. This protects against inadvertent writes during power up or brown out conditions.

Best Practices

nvSRAM products have been used effectively for over 15 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 an nvSRAM are programmed on the test floor during final test and quality assurance. Incoming inspection routines at customer or contract manufacturer’s sites will 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, etc. should always program a unique NV pattern (e.g., 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.

Power up boot firmware routines should rewrite the nvSRAM into the desired state (autostore enabled, etc.). 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 (program bugs, incoming inspection routines, etc.).

If AutoStore has been firmware disabled, it will not reset to “autostore enabled” on every power down event captured by the nvSRAM. The application firmware should re-enable or re-disable autostore on each reset sequence based on the behavior desired.

The VCAP value specified in this data sheet includes a minimum and a maximum value size. Best practice is to meet this requirement and not exceed the max VCAP value because the nvSRAM internal algorithm calculates VCAP charge 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 with Cypress to understand any impact on the VCAP voltage level at the end of a tRECALL period.

Low Average Active Power

CMOS technology provides the STK14D88 with the benefit of power supply current that scales with cycle time. Less current will be drawn as the memory cycle time becomes longer than 50 ns. Figure 13 shows the relationship between ICC and READ/WRITE cycle time. Worst-case current consumption is shown for commercial temperature range, VCC = 3.6V, and chip enable at maximum frequency. Only standby current is drawn when the chip is disabled. The overall average current drawn by the STK14D88 depends on the following items:

The duty cycle of chip enable

The overall cycle rate for operations

The ratio of READs to WRITEs

The operating temperature

The VCC level

I/O loading

Document Number: 001-52037 Rev. **

Page 12 of 17

[+] Feedback

Page 12
Image 12
Cypress STK14D88 manual Software Recall, Data Protection, Best Practices, Low Average Active Power