• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 82 subscribers
  • Views 134 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TMS320F280039: Inquiry on possible effect of wave soldering to Flash Memory

Alfred Logico
Expert 4570 points
Part Number: TMS320F280039

Hi Team,

Good day!

Customer just reported to me today from their factory that they are facing concern regarding F280039CSPM MCU flash memory.

There were instances where the firmware loaded onto the MCU flash memory either becomes erased or the data in a section becomes altered.

  • The MCU programming is done after AI SMT stage, prior to MI wave solder.
  • MCU is de-energized after initial programming, then 3.3V is re-applied and flash memory contents checked.
  • The board is then subjected to wave solder, with a maximum solder temperature of 2630C.
  • The assembled unit is then loaded to T1 station, where in some instances it would not power-up or communicate at all due to MCU “program corrupted”.

Kindly advise on any potential effects on MCU flash memory when subjected to up to 2630C wave solder temperature environment.

The only item I found on the datasheet is possible INTOSC frequency shift due to reflow (section 6.12.3.5 page 97).

Thanks!

BR,

Alfred

  • 0
    TI__Genius 10291 points

    Hello Alfred,

    Although the peak temperature (263C) is close to our data sheet max (260C -- see the packaging information tables at the end of the data sheet), we usually expect customers to use a regular JEDEC-compliant IR reflow process. QFP packages with a pitch this small (0.5mm) are likely not suitable for wave soldering/full solder immersion. Please refer to JEDEC JESD22-A111. I would strongly recommend IR reflow instead.

    Best regards,
    Ibukun

  • 0 in reply to Ibukun Olumuyiwa
    TI__Expert 4570 points

    Hi Ibukun,

    Thanks for the feedback! Will advise customer to do IR reflow instead, but to check, is there a possibility that the flash memory can be affected by the high temperatures?

    Thanks!

    BR,

    Alfred

  • 0 in reply to Alfred Logico
    TI__Expert 4570 points

    Hi Ibukun,

    Here is feedback from customer:

    To clarify our setup:

    • Step1: F280039CSPM is reflow soldered to main board along with other SMT parts.
    • Step2: The appropriate firmware is flashed onto the MCU, and image is verified again afterwards.
    • Step3: The manual insert parts are populated onto the main board, which then undergoes selective wave solder.
      • For selective wave solder, the F280039CSPM is covered to prevent direct contact with the solder.
    • Step4: The PSU would undergo initial testing.
      • It is during this step that the problem is observed – PSU does not power-up at tester, and upon bench verification the MCU flash memory is observed to be blank or there is mismatch with the originally flashed firmware.

    The SMT components, including the F280039CSPM, is covered by a thin board / fixture to prevent direct contact with wave solder.

    There are holes or cut outs on the fixture to allow solder to contact the selective locations on the PCB. Unfortunately I can’t find any photo of this.

    The F280039 which had erased flash memory can still be reflashed. The PSU would then operate normally after MCU reflash.

    Thanks!

    BR,

    Alfred

  • 0 in reply to Alfred Logico
    TI__Genius 10291 points

    This may depend on the high temperature exposure profile. Exposing the device to temperatures exceeding the peak solder temperature, particularly for an extended time (or cumulative exposures) could result in device degradation. I would refer your customer to this document: www.ti.com/lit/pdf/spraby1a.

    Also note at the end of that document that it says that wave soldering for components other than SOIC or through-hole is not recommended.

  • 0 in reply to Ibukun Olumuyiwa
    TI__Expert 4570 points

    Hi Ibukun,

    Customer wants to clarify on your statement: "Exposing the device to temperatures exceeding the peak solder temperature, particularly for an extended time (or cumulative exposures) could result in device degradation. "

    • What kind of degradation would be induced by slightly exceeding the peak solder temperature?
    • Does this mean that it is not advisable subject the MCU to high temperatures (ie: wave solder 260C) after programming?

    This would help guide customer on the proper corrective actions to avoid repeat of “MCU No Program” failure at our factory.

    Thanks!

    BR,
    Alfred

  • +1 in reply to Alfred Logico
    TI__Genius 10291 points

    Hi Alfred,

    I am not able to define exactly what type of degradation can happen beyond the peak solder temperature. This type of environment is not part of TI's QFP package qualification processes. I would agree with the second statement -- if wave soldering must be used, it should be done before programming the MCU especially since corruption is being observed. But overall, TI has not qualified this package for wave soldering; the customer will need to evaluate this risk for themselves in their assembly flow.

    Best regards,
    Ibukun

  • 0 in reply to Ibukun Olumuyiwa
    TI__Expert 4570 points

    Hi Ibukun,

    Thanks for the very helpful feedback! I will let you know if customer has any further questions.

    Thanks!

    BR,

    Alfred