This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMS320F28375S: SRAM number of rows and process node

Part Number: TMS320F28375S


We are trying to assess the ability to detect bit state changes due to soft errors using the ECC and parity protection mechanisms built into the microcontroller.  In this case, the kind of soft errors I'm referring to are described in section 3.4.1 of SPRACC0.

The combination of sections 2 and 3.4.1 of SPRACC0 describe the physical layout of the SRAM and how close any two bits in a word may be near each other.  In order to assess the susceptibility of a single charged particle affecting multiple nearby bits, what are the number of rows in the SRAM array and the process node used in this memory?  I have not been able to find this information in any literature and it would be very helpful to understand what additional software protections are needed to detect changes.

  • Hi Jordon,

    All SRAM's on TMS320F28375S are either protected by ECC or Parity as diagnostic. Refer to "Table 6.6 Memory Types" in device datasheet.

    Parity are effective to detect single bit errors and ECC is effective to detect 2-bit errors and correct single bit errors (SECDED).

    Any single soft error event to escape detection by Parity should be resulting in Multi-bit-Upsets with at least 2-adjacent-bit disturbed. Most data SRAMs on TI's 65nm node are designed with mux factor sufficient to avoid the Multi-bit upsets resulting in very rare event(MBU is <1% of total SER). ECC provides additional feature to correct the single bit errors to improve the availability of the system.

    Regards,

    --Ashish