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.

MSP430F6746: Questions about Flash of MSP430F6xx

Expert 2780 points
Part Number: MSP430F6746

Hi,

Does this MSD430F6746 flash architecture refer to this application note?


MSP 430 Flash Memory Characteristics
Http://www.ti.com/lit/pdf/slaa334

Regards, DA

  • Hi DA,

    The app note you are referring to is quite old and therefore only mentions the legacy MSP4301xx/2xx/4xx devices, but the technology explained does apply to all MSP430 flash devices. Do you have a specific question regarding this app report?

    Regards,
    Ryan
  • Hi,

    There is a problem in the customer's board.

    The contents of Flash has changed.
    A bit in Flash has changed from 0 to 1.
    (Region not program)


    Here was a question from the customer.

    How does Flash's contents change when the charge is removed in Flash memory?
    "0" to "1"  or  "1" to "0"?

    I decided the answer from "0" to "1".
    Is that understanding?

    Regards, DA

  • As shown in section 2 of the application report, in an erased (1) cell, the floating gate has a positive charge, and in a programmed (0) cell, the floating gate has a negative charge.
    However, according to Wikipedia, an erased cell is uncharged, so I guess the application report means positive compared to the programmed state.

    Anyway, section 3.1.2 says that

    single bit failures due to leakage currents show erroneous 0s that should be 1s. These single bit failures based on leakage currents can never read erroneous 1, instead of expected 0s.

    There does not appear to be a hardware error mechanism that would result in a 0 → 1 error.

  • As Clemens mentioned, single bit failures result in erroneous 0s that should be 1s (from 1 to 0), not the other way around. A bit will change from 0 to 1 if it has been erased but should not occur from leakage current or hardware malfunction.

    Please further describe the application and system environment in which this issue occurred. Has only one device displayed this problem so far? For how long has this device been active? Is it possible that the memory area's flash endurance has been exceeded?

    Regards,
    Ryan
  • da said:

    There is a problem in the customer's board.

    The contents of Flash has changed.
    A bit in Flash has changed from 0 to 1.

    There is marginal read for 5xx/6xx flash devices, for checking if device was flashed correctly.
  • Hi,
    Is it certain that this problem (0 -> 1) does not occur?

    This flash memory problem occurred after 8 months or more worked.
    The area containing this bit is not rewritten in the application.

    The same problem has been confirmed with 2 devices.
    A lot of devices are not experiencing this problem.

    Regards, DA
  • > an erased cell is uncharged, so I guess the application report ....

    If it is a programmed bit, what if it is charged?

    The bit in which the problem occurred is not the erased bit (1) but the programmed bit (0).



    Regards, DA
  • The bit in which the problem occurred is not being rewritten.

    The occurrence is the area used as a program.
    It turned out that a problem occurred in the flash memory because the program that worked normally malfunctioned.

    Regards, DA
  • Hi DA,

    Going from 0->1 is more unusual as you should only be able to do this with an erase operation. Common flash corruption causes and best practices are listed in the following Wiki: processors.wiki.ti.com/.../MSP430_Flash_Best_Practices

    You should also check the Vcore level vs operating frequency and make sure that the levels are stepped up one-at-a-time. Vcc/MCLK violations have caused flash corruption in the past. Additionally, is the SVS enabled and set correctly to make sure the part will catch if Vcc starts to drop below the minimum required level for the Vcore setting used? The following threads might also have some useful information:

    e2e.ti.com/.../
    e2e.ti.com/.../19071

    Regards,
    Ryan

**Attention** This is a public forum