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.

Tiva Xm4C129ENCPDT no alternate UART0 pin

Other Parts Discussed in Thread: TM4C129ENCPDT, TM4C1294NCPDT

 TM4129 remain as XM with no alternate to CAN0 R/TX colliding with UART0 used for debug?

 I found pin assignment as horrible, every peripheral is spread in a hard to route pattern, worst is EPI taking signals from each of 4 side of ic package, this result of uneven length equalization. Also USB and LAN can be better in a different package like tqfp 144, with  some pin NC near these interfaces can help tune impedance rule from IC pin to socket.

  • Roberto,

    Thank you for the feedback. Our layout best practices are documented in the TM4C129x Design Guide

    I find the easiest way to identify options for the pinout is using our pin mux utility.

  • miguel2667 said:

    Our layout best practices are documented in the TM4C129x Design Guide

    ----- options for the pinout is using our pin mux utility.

     Miguel... I am using pinmux, I read thoroughly design guide, I am near test of prototype BUT...

     CAN0 RX, TX as CAN1RX TX

     simply CANNOT be assigned to nothing else and CAN0 collide with the UART0 again no alternate assignment...

    Other family member 123x have from 2 to 3 alternate pin selection for can, this one has not..

     UART0 is the debug console I need both CAN so I miss debug console.

     PA0, PA1 must be reserved to debug console and peripheral sharing pin IMHO MUST have at almost one alternate.

  • If debug can only happen on UART0 then there should be nothing else on these pins. And certainly, if anything is put on these pins, it should also be available elsewhere.

    BTW, I found the pin mux utility useless. The utility should be redone so that a user types in how many UART, SSI, I2C, GPIO etc they want and the utility should juggle things around to try to make it happen. It should also leave the most versatile pins as unused so that these can be brought out to an expansion header and potentially utilised with maximum versatility at a future date.

    I had to do all my pin assignments by hand for a TM4C129ENCPDT. I included a 40 pin Booster Pack expansion connector and made sure the most versatile pins went to the connector. Fortunately my application does not use CAN (which with only 2 CAN ports and no reassignment is not good).

    TI, is there any way to put debug on one of the other 9 UARTs?

  • Roberto,

    Are you using one of our dev boards or making your own?

    If creating your own board simply use a different UART port for debug (UART1,2,3 etc).

    If you're using the TM4C129x Development Kit then you are in luck, CAN0 also is broken out on PT0 and PT1. As we are currently only making one kit with a snowflake part on it this solves your problem.

    Regards,

    -Austin

  • Vito Casa said:
    I had to do all my pin assignments by hand for a TM4C129ENCPDT. I included a 40 pin Booster Pack expansion connector and made sure the most versatile pins went to the connector. Fortunately my application does not use CAN (which with only 2 CAN ports and no reassignment is not good).

     Hi Vito, thank for parallel support ;) I see you are using my same part (not too many are available till now), how you solved the impedance contraint of both LAN and USB? alse these are in a worst position with XTAL too close, A more clever pinout on TQFP can help routing, PCB can be routed on a 2 layer but pin distribution is so strange and I switched to 4 layer, most of board is routed on layer 1 with layer 2 solving the impedance constraint posed by LAN and USB, I found no way to maintain an uncrossed solid ground plane on bottom layer....  I also found bad this fine pitch case, impedance mismatch is too far from 90/100Ohm needed. BGA version is more clever .

  • Hi Robert, I'm using the  TM4C129ENCPDT since it appears to be the only  TM4C129x in a TQFP package that is currently available. Ultimatly I will switch to the very similar TM4C1294NCPDT since we don't need the AES/DES/SHA/MD5 functionality and presumably the TM4C1294NCPDT would be cheaper since it has less functionality.

    I'm currently at the schematic capture stage (with some design still to go) so have not yet considered layout consideration such as getting the USB and Ethernet track impedances right. Thanks for the heads up, it's not something I've thought about yet. I've already printed out a copy of "System Design Guidelines for the TM4C129x Family of Tiva™ C Series Microcontrollers", so will be having a good read before the PCB layout proceeds. If you have any tips that would be great!

    In fat, I may have to talk to you about other aspects. I've been away from microcontroller design work for about 20 years! Last time I did any of this sort of design was using  8051 microcontrollers. Things have certainly changed over the decades. Much more exciting now!

    Regards, Vito

  • Vito Casa said:
    I'm currently at the schematic capture stage (with some design still to go) so have not yet considered layout consideration such as getting the USB and Ethernet track impedances right. Thanks for the heads up, it's not something I've thought about yet. I've already printed out a copy of "System Design Guidelines for the TM4C129x Family of Tiva™ C Series Microcontrollers", so will be having a good read before the PCB layout proceeds. If you have any tips that would be great!

     Hi Vito, I got away for  a long while, I am troubled by too much things, design proceed slower and I count to finish first layout by the  end of January. I also found another part with same issue on CAN1, the processor mounted on LaunchPad has 3 alternative for can0 but again CAN1 is firmly hardwired to USART0, the default port needed by debug and bootloader too...

     About Impedance please tell me how can I help you, as I wrote no chance to have a solid plane with 2 layer PCB so i switched tp  4 layer, again 4 layer is more simple to route with BGA, I don't know if continous vibrations tend to generate mechanocal fatigue, solder ball stress or structural damages(die or ceramic crack).

     I am now at part collection, when all part are successfully collected I start manufacturing of prototype.

  • Hi Roberto,

    So, are you using the BGA or TQFP package for your product? 

    I've found some TI documentation about routing USB and Ethernet connections and I think I'm OK for now. We will be using a four layer board which will make life a lot easier.

    How does availability of TM4C129ENCPDT affect you?

    Regards, Vito