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.

TMS320F28035: TMS320 : Compatibilty between several C2000 microcontrolers

Part Number: TMS320F28035
Other Parts Discussed in Thread: TMS320F28075, TMS320F28379D, LAUNCHXL-F28379D, TMDSCNCD28379D, TMS320F280049C

Hi,

In my company we are using a TMS320F28035, but we are thinking to change it for a bigger one. We looked at the TMS320F28075 to increase the memory space, but no evalboard is available. So we decided to look at the TMS320F28379D because an evalboard is available everywhere.

We have several questions about these products :

  1. Is that hard to disconnect One CPU and One CLA in the F28379D in order to simulate a F28075?
  2. Is the code based on F28035 portable to a F28075 or a F28379D (if we don't take in account the pining and drivers)?
  3. We saw that both F28075 and F28379D are available in a 176-pins package. Are they pin-to-pin compatible?

Best regards

  • Alexis,

    The F28379D is a natural superset of the F28075. 

    1. It is very simple to develop code and limit your usage to one CPU and one CLA. We provide many examples that utilize only one CPU and CLA. You don't even need to load code on CPU2. unless you choose to wake it up to use, it will be in standby mode and not executing. You can also manipulate your linker command file to limit the available memory for your application to use. 
    2. Many of the peripherals are similar between the F28035 and F28075. There will be incremental improvements to some peripherals (SPI, EPWM, CLA, etc) and there will be new or different modules all together (eCAN vs CAN, XBAR, IPC, etc). 
    3. Yes. the 176 pin packages are pin-to-pin compatible.

    I recommend getting your hands on either the LAUNCHXL-F28379D or the TMDSCNCD28379D and start developing and porting. You might also consider looking at the TMS320F280049C as an option as well this is slightly different than the F28075. Please refer to the C2000 Peripheral Reference Guide: http://www.ti.com/lit/spru566

    If you are interested in getting started with the F28379D , refer to the Multi-Day workshop: http://processors.wiki.ti.com/index.php/C2000_Multi-Day_Workshop 

    Thanks,

    Mark

  • Alexis,

    You can also take a look at the new F28004x devices, which fall in between F2803x and F2807x.

    -Tommy
  • Hi Tlee,

    thank you for your answer. The F28075 really was our less powerful microcontrolers in our list of possibilities.

    The opportunities of evolution are greater with the F28379D so we decided to take this one.

    One more time thank you.

    Regards
  • Hi Mark,

    thank you for your return. We finaly took the F28379D, thanks to your answer, for the opportunities of evolution that this microcontroler is offering.

    One more question, we saw that the F28379D has Can modules instead of eCan in F28035. The can is an important feature of our product. What are the exact differences between the Can and the eCan module (hardware or/and software)?

    Regards
  • Alexis,

    The F28379D is a fine choice. It is our most powerful MCU to date and is scalable to many lower cost alternatives to fit any any application.

    These modules are very different and not code compatible. the CAN on F2837x devices is the module C2000 is moving forward with. I will pull in a CAN expert and he can perhaps provide a little more insight to the specific differences.

    -Mark
  • Alexis,

                Following are the 2837x DCAN features/ enhancements compared to eCAN:

     

    • Parity check mechanism for all RAM modules
    • Automatic Retransmission (upon loss of arbitration) can be disabled.
    • Silent mode (Node listens passively)
    • Mailbox RAM may be combined to form FIFO buffers.

     

    Differences compared to eCAN:

    • 28xx eCAN register map/bit allocation/bit definitions are not compatible with DCAN.
    • DCAN follows a different programming model (in terms of how mailbox RAM is accessed and also in terms of the header files provided by TI).
    • eCAN time-stamping feature not available.
    • Programmable transmit priority not available.
  • Hareesh,

    Thank you very much for these clarifications. An estimation of the software modifications which are involved is required on our part.

    I left this thread opened if we have questions in the following days.

    Regards,

    Alexis