• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 82 subscribers
  • Views 367 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMS320F28379D: Position Manager : processing time

Eileen Becker
Prodigy 20 points
Part Number: TMS320F28379D


Hello everyone,

I`m new to the TI community so please excuse possible mistakes in this post.

I`m working with the F28379D MCU and trying to run a bissc tranfer using the Position Manager library.

The objektive is to run the complete interrupt with the bissc transfer in less than 5 us.

The problem is that some of the TI library functions are taking quite a while. (e.g. PM_bissc_SetupNew SCDTransfer() needs 742 CPU cycles which result in 3.71 us at 200 MHz Clk)

All together the interrupt currently takes about 9 us.

Is it possible to get the C-files for the PM_bissc library to maybe delete some ramifications that aren`t needed in my case?

Or has anybody an other advice?

Thanks in advance

Eileen

  • 0
    TI__Mastermind 24095 points

    Eileen,

    Did your measurements match the CPU cycle numbers quoted in the Library document? It seems to match the expected cycle count as per the user guide.
    At this stage we are using a library based approach for PM functions and hence will not be able to provide the source code.

    -Bharathi.

  • 0 in reply to Subrahmanya
    Prodigy 20 points

    Bharathi,

    Thanks for your quick reply, yes my measurements and the data in the guide match pretty good.

    Do you by chance know a faster way to realise the communication?

    At a trade fair one of your colleague told us, that we could solve the problem with the chip and the PM_ bissc_library.

    Currently we are using an other controller and additional components for the transmittion. 

    An other possibility would be, to think about an own way to realise the communication on the chip, but that would presumably take a lot of time to get it working.

    Eileen

  • 0 in reply to Eileen Becker
    TI__Mastermind 24095 points

    Eileen,

    Good to learn that it's matching the expected numbers specified in the user guide.
    Beyond this further optimization specific to the encoder/interface you are using is not possible as these functions are packaged as pre-compiled libraries and hence you'll not be able to make changes at your end.

    -Bharathi.