• State
  • Replies 3 replies
  • Subscribers 79 subscribers
  • Views 82 views
  • Users 0 members are here
Support feedback
Options
Options
Related

Original question:

MSP430F2370: Wrong DCO constants after flashing

MSP430G2744: The values ​​0x10F0-0x10FF in Segment A become 0xFF (Blank).

O.H
Mastermind 6370 points
Part Number: MSP430G2744
Other Parts Discussed in Thread: UNIFLASH

Tool/software:

Hi experts,

We're currently investigating an issue very similar to the one in the original thread.

After mounting the MSP430G2744IRHA, the "FET-Pro430" software displays a "Fail" status when writing firmware. Specifically, the values ​​of 0x10F0 to 0x10FF in Segment A are 0xFF (Blank).
Our goal is to determine the cause of the values ​​0x10F0-0x10FF becoming 0xFF (Blank).
Could you please share TI's position on the following points?

Q1: Is there a possibility that data in this area will be erased due to a power surge when power is supplied to the CPU? Please let us know if you have any similar cases.

Q2: Is there a possibility that data in this area will be erased if the circuit deviates from the recommended circuit in the MSP430 Hardware Tools User's Guide (Rev. AH)? Please let us know if you have any similar cases.

Q3: Is there a possibility that data in this area will be erased during mounting or reflow? Please let us know if you have any similar cases.

[Details]

  • This issue only occurred in one lot (250 units), affecting 22/220 units.
  • Reading 30 units from Uniflash using the ICs individually did not return 0xFF (Blank). 
    We are currently requesting implementation and programming of the above 30 units.
  • The circuit and tools used have not changed in nearly 10 years.
  • No significant deviations from the recommended circuit in the MSP430 Hardware Tools User's Guide (Rev. AH) (minor differences in resistor and capacitor values).
    Figure 2-2. Signal Connections for 2-Wire JTAG Communication (Spy-Bi-Wire) Used by MSP430F2xx, MSP430G2xx, and MSP430F4xx Devices
  • Recalibration is possible using "msp430g2x44_dco_flashcal.c."
  • The "FET-Pro430" "Memory Option" does not have any effect.

Best regards,
O.H

  • 0
    TI__Genius 15732 points

    Hi O.H,

    The area that you are trying to modify in the information memory area. Normally, this area should be left alone unless there's a good understanding of how to modify the values here.


    1. I haven't heard of something happening to our devices. The only instance I can think of this happening is if, for some reason, the device was damaged
    2. Similar to the answer above, if the device is damaged, the behavior will be abnormal. Personally, however, I haven't heard of such a case before.
    3. Heat could be a cause for the device to be damaged, especially if it surpasses the recommended operating temperatures. However, heat normally doesn't cause data in FLASH to change (unless it was a FRAM device since FRAM is sensitive to heat.)

    Best Regards,

    Diego Abad 

  • 0 in reply to Diego Abad Sajamin
    Mastermind 6370 points

    Hi Diego Abad

    Thank you for your supports.

    Diego Abad Sajamin said:
    1. I haven't heard of something happening to our devices. The only instance I can think of this happening is if, for some reason, the device was damaged
    2. Similar to the answer above, if the device is damaged, the behavior will be abnormal. Personally, however, I haven't heard of such a case before.
    3. Heat could be a cause for the device to be damaged, especially if it surpasses the recommended operating temperatures. However, heat normally doesn't cause data in FLASH to change (unless it was a FRAM device since FRAM is sensitive to heat.)

    Based on the answer above, it seems unlikely that the failure is due to a hardware issue.

    My statement may have been a little misleading, so I'd like to provide the following additional information.

    • The error log when the writing failed is below.
      "DCO constants verlifiacation... failed
    • The IC that failed to be programmed was taken from the reel and installed.
    • 220 ICs on the reel have been installed, and the data for the remaining 30 ICs was confirmed using Uniflash.
    • There is no content in the firmware that manipulates 0x10F0-0x10FF.

    Q4: Even with the above information, do you still think the possibility of a hardware-related failure is extremely low?

    Q5: Have there been any reports of bugs with the "FET-Pro430"?

    Q6: Could you tell me how the problem in the original thread was resolved? I don't mind answering in private chat.

    Best regards,
    O.H

  • 0 in reply to O.H
    TI__Genius 15732 points

    Hi O.H.,
    Q4. Actually, I think the opposite. There might be a posibility that what the device showing the issue could have been damaged, though I'm not sure exactly how it could have damage.

    Q5. I haven't heard of any in particular, nor I can't find any previous instances in E2E that may relate to it. However, I would recommend browsing in E2E just in case I missed something.

    Q6. Most likely what happened in the original thread is that it automatically got closed after some time. However, I would recommend the same as the original thread on regards of comparing the sections between a good and a bad unit. 

    Best Regards,

    Diego Abad

**Attention** This is a public forum