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.

TMDSCNCD28069MISO: TMDSCNCD28069MISO and absolute encoder feedback via RS-422

Part Number: TMDSCNCD28069MISO
Other Parts Discussed in Thread: DESIGNDRIVE

hi,  i'm wondering how to connect RS-422 from motor's absolute encoder to control card.  RS-422 has motor position data which is 16 bit and up to 1Mbps data rate.  this goes through a TI isolated transceiver and then ... to which pins of the TMDSCNCD28069MISO control card?  I thought it should goto SCI-A GPIO 28/29 which is pins 50/93 on control card connector.  what i'm worried is that the SCI on control card can not keep up with the motor position data, especially during high motor acceleration. 

  • Hi,

    Which encoder are you using? Most of these encoders have their own communication protocols (Ex: EnDat, BiSS-C, Tamagawa.. etc.).
    So, it depends on the type of encoder you are interfacing.
    Also visit the page below to learn more about the position manager solution - which mainly is intended to interface different types of encoders.
    http://www.ti.com/tool/DESIGNDRIVE

    -Bharathi.

  • true the absolute encoder has other options but it was not selected by me, so for sure (unless you say it can't be done) the motor position feedback is coming on RS-485 to the control card.  the data rate is selectable, the maximum is 1Mbps which honestly seems too slow for real time motor control of a large high performance drive.. and it has 200us latency :(

    the link you sent talks of the Dalphino processor which is different than the piccolo but more to the point DesignDrive only talks about incremental encoder, sin/cos, resolver, basically everything else other than absolute encoder.  was not that helpful unfortunately

  • Hi,

    The PositionManager solution for absolute encoder interfaces is available on Delfino family. Same solution can't  be adapted to 28069.
    Data rate of 1 Mbps is not so uncommon in case of encoders, it really depends on you position loop frequency.
    I'm not sure about the encoder in use in your system and the data format/protocol - to confirm whether or not it can be achieved, using 28069.

    -Bharathi.

  • ah, i see.  RS-422 is just the "form" that the position data protocol will take, but it might actually be using enDAT, BiSS or some other. 

    https://www.rls.si/products/aksim-off-axis-rotary-absolute-encoder

    encoder manufacturer is Renishaw type RLS.  from the manual "The encoders come with SSI, SPI, PWM, BiSS, asynchronous serial RS422 and USB communication interfaces".  i'm expecting the motor position data to be coming by asynchronous series RS422.  the data rate has max of 1Mbps but i think the control card SCI port is max 200kbps so this is why i wasn't sure if solution would work... seems the data is pretty slow for electric vehicle application. i planned to use isolated RS-485/422 transceiver and then mapping the receive and transmit to x69 control card SCI port  (which i think was GPIO 28/29).  my feeling is that of the available options from encoder, BiSS will not be compatible with the x69 control card .. which i guess is OK for me cause i was not expecting to use that anyway but rather just straight asynchronous series RS422.

    hopefully my explanation is not too confusing

  • Though the physical layer is RS422 - you may not be able to directly hook up to the SCI interface, just due to the data format. Nevertheless, you can try it out. If it's a very slow interface, you could try GPIO bit bang option in s/w, instead of SCI port.
    Note that the SSI and BiSS are synchronous in nature - so the interface needs a clock.
  • hmm.. asynchronous serial RS422, from the encoder datasheet data format is "8 bits, no parity, 1 stop bit" which seems pretty standard.  no problem for SCI here.  the encoder datasheet goes on to say "Data is transmitted MSB first; big-endian order"  i don't know interface of SCI well enough off top of my head but i imagine this is not a show stopper either.  the data packet is 16bit which is OK for control card SCI.  so all seems OK other than i wish i could get control card to accept data at 1mbps.  from the first read i made of the controlcard F28069 it seems like 200kbps was max. 

    my next question.. from using control card TMDSCNCD28069MISO, i plan to use JTAG for programming and debug.  simultaneously i plan to use both SCI-A and SCI-B for DIMM connector communication with outside world (via isolation transcievers).  SCI-A will be dedicated to motor position feedback coming in on RS-422 as previously discussed.  SCI-B is for bi-directional communication with a different uC by RS-485.  looks like there is some issue with this control card to use JTAG and SCI-A by DIMM connector at the same time?  seems JTAG makes use of SCI-A  ... which means my plan is not going to work?

  • The FTDI chip does just take the GPIO28 and GPIO29 as UART and interfaces them with the USB so that you can access the UART from the PC without any extra connection. If you do not connect anything to the USB of the XDS100, you can still use the GPIOs for MCU to MCU communication through the base board. If you never intend on using the routing through the FTDI chip, you can disconnect them by cutting the traces (if there is space to do so. I don't have a board directly in front of me).

    I don't see a problem with using both channels how you intend to even without modifications.
  • A colleague of mine just mentioned to me that there is a switch on the ControlCard (SW3) that will disconnect the SCIRXA signal from the isolator. This will prevent any contention that may occur from multiple drivers on the RX signal allowing you to use SCIA how you intend to.