• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 83 subscribers
  • Views 581 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.

TMS320F28388D: EMIF Address line A11 and data line D9 not available on HSEC180 Docking Station

Mark Keranen1
Prodigy 110 points
Part Number: TMS320F28388D
Other Parts Discussed in Thread: TMDSHSECDOCK, TMDSCNCD28379D, TMDSCNCD28388D

I am attempting to use the EMIF1 interface on the TMS320F28388D paired with the TMDSHSECDOCK to interface with an external parallel nor flash chip.

I am accessing the EMIF interface signals via the header pins on the TMDSHSECDOCK docking station, but it seems that not all address and data pins are accessible. Particularly EMIF1_A7 and EMIF1_D9.

EMIF1_A7 uses GPIO47 or GPIO51. GPIO47 is not available on the TMDSHSECDOCK, and GPIO51 is used by EMIF1_A11 which has no other mux options. This leaves me not able to use both A7 and A11 address lines in the interface.

EMIF1_D9 (GPIO75) is simply not present on the TMDSHSECDOCK so I have no way to connect this data line to the flash memory component.

Is this an oversight that a couple of signals are missing on the interface or do I not have the correct diagram for the headers?

  • 0
    TI__Guru** 113565 points

    Hi Mark,

    On this device we have EtherNet and EtherCAT as new interface which are also high pin count interface and focus was to make these available on docking station hence EMIF pins may not be available. I am looping in our board design expert to provide any further input on this.

    Regards,

    Vivek Singh

  • 0 in reply to Vivek Singh
    Prodigy 110 points

    Is there any way I can work around this? Custom GPIO to make up for the missing pins?

    Otherwise, is there a better option available with all available EMIF pins?

  • 0 in reply to Mark Keranen1
    TI__Mastermind 42195 points

    Mark,

    TMDSCNCD28379D should have better access to the EMIF connections, and its EMIF is very similar to the F28388D from a prototyping perspective. That may work well for your purpose.

    Speaking specifically about TMDSCNCD28388D we did consider this during the design phase and as Vivek said we proitized enabling both Ethernet, EtherCAT and our standard controlCARD "control functions". All of these constrains made it difficult to support EMIF.

    Nevertheless I believe EMIF1 does have all of the necessary pins to function in 8 bit mode. As you pointed out its that pesky D9 which causes an issue with 16b mode. The configuration I have shown below does not enable the "wait" function.

    If you are really interested in 16b mode, you may be able to add a "blue wire" to bring GPIO 75 down from the Ethernet TX_D0 line to an unused pin in the edge connector. This would limit the speed of the interface, but could still work good for proving out your software.

    Regards,
    Cody

  • 0 in reply to Cody Watkins
    Prodigy 110 points

    Thanks,

    I can get by in 8 bit mode in that configuration but I don't have access to the full address space I need since I can't access all the address pins. Any workarounds for that?

    Specifically I lose the ability to access A11 when I have to mux A7 to GPIO52 (since GPIO47 isn't on the header) which removes the ability to use A11 since it doesn't have other mux options.

  • 0 in reply to Mark Keranen1
    TI__Mastermind 42195 points

    Mark,

    that is correct, you do not gain access to the entire address space.

    One workaround for this is to use the input and output XBars to move the A7 signal to another pin. this will allow GPIO51 to be open for A11. Setup the GPIO47 for EMIF, connect the input XBAR to this channel, and then connect an output XBAR channel to this input xbar channel.

    Please note that this will add some latency to the signal and will limit how fast you can operate your EMIF interface, but for experimentation purposes it should work.


    Regards,
    Cody