• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 82 subscribers
  • Views 584 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.

Original question:

TMS320F28335: SCI communication uses address bits to wake up abnormally

TMS320F28335: The connection between SCI Multiprocessor Communication modes and baud rate

Weng
Prodigy 40 points
Part Number: TMS320F28335

Hello,

         SCI Multiprocessor Communication has  two  modes , one  is Idle-line Multiprocessor Mode, the  other is Address-Bit Multiprocessor Mode.As the baud rate increases, there are different results for the above two modes. I increased the baud rate to 460800, and can communicate normally in Idle-line Multiprocessor Mode. However, the  Address-Bit Multiprocessor Mode did not  work properly. Even when the baud rate was 115200, the address bit pattern can only work by adjusting the baud rate error.Therefore，I have the following questions.

1. Is the Address-Bit Multiprocessor Mode can only be used when the baud rate is lower than 115200? In contrast, Idle-line Multiprocessor Mode can use more higher baud rate, including a baud rate of 460800.

2.As mentioned in the F2833x Technical Reference Manual,the address-bit format is typically used for data frames of 11 bytes or less. The idle-line format is typically used for data frames of 12 bytes or more. I want to know the specific causes of this phenomenon.

Thanks and best regards,

Weng

  • 0
    TI__Genius 17315 points

    Weng,

    Let me investigate your first question about the two SCI communication modes and get back to you.

    To answer your second question - 'address-bit format is typically used for data frames of 11 bytes or less; and the idle-line format is typically used for data frames of 12 bytes or more'.  As you know, with address-bit mode there is ALWAYS an extra bit and the idle time is of no significance, but with idle-line mode there is NO extra bit and an idle period of 10 bits (or more) is required to separate the blocks.  This statement simply refers to the break-even point of efficiency in terms of bytes when comparing the two communication modes.  That is, there is a point where including the extra bit is offset by the required idle period of 10 bits.  Let me know if this answers this question.

    - Ken

  • 0 in reply to Ken Schachter
    TI__Genius 17315 points

    Weng,

    For your first question, it is not unusual for each device to vary in terms of the baud configured error (target baud vs. actual configured baud).  For example, on one device family (e.g. F2833x) 115200 might work well but on another device family (e.g. F2806x) it might not work well.  However, I am not sure why there is a difference between address-bit mode and idle-line mode.  As far as we know, there are no limits on address-bit mode.  Does the address-bit mode have more bits in the packet?  If so, that would make it less robust to baud-rate mismatches.

    - Ken

  • 0 in reply to Ken Schachter
    Prodigy 40 points

    Ken，

    For the address-bit mode, the data packets we send  will not exceed 10 bytes during the test . But in reality, there will be an exception if it exceeds 2 bytes.Since the difference between  address-bit mode and  idle-line mode can't be determined, then I will not delve into this question. After all, the idle bit pattern is working fine now.

    - Weng