• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 101 subscribers
  • Views 569 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.

TMS320C6657: Timing for EMIF nor flash interface, hold time constraint not met.

paul kerstetter
Intellectual 455 points
Part Number: TMS320C6657

I was performing timing analysis on the RC28F128P30B65 Nor flash and the TMS320C6657 using the EMIF interface.  To me it looks like this NOR does not work due to timing parameter denoted as R10 (Output hold from first occurring address, CE#, or OE# change)  min 0 nS in the data sheet.  Also see figure 26 in the data shee (Asynchronous Read-to-Write Timing).  This is because in the ti emif timing specifications th(OEH-D) (Input hold time from OE high to D invalid) is a minimum of 0.5 nS.  See table 5-17 EMIF16 Asynchronous Memory Timing Requirements and figure 5-20 below.

Is there any way around this besides using external parts?

 see this link for nor data sheet download:

 

  • 0
    TI__Guru**** 161600 points
    Hi Paul,

    I've notified the design team. their feedback will be posted here.

    Best Regards,
    Yordan
  • 0
    TI__Mastermind 42545 points

    Hi Paul,

    There isn't a way to program the EMIF interface to compensate for that 500ps hold time requirement on the data input. Most customers will use the delay associated with the PCB to achieve the hold requirement. Remember that the timing is measured at to the pins of the C6657. Once the OE signals is launched there is some delay between when the signal is valid at the pin of the C6657 and when the signal arrives at the memory component. There is additional delay for the data signals to return. As you increase the length of the OE and data lines, you increase the hold time on a read access. 

    Regards,

    Bill

  • 0 in reply to Bill Taboada
    Intellectual 455 points

    Thank you,  as a side note the i ofund out the memory part is going obsolete.  Switching to a new memory now.