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.

AM3359 ICE etherCAT full feature appl -- state machine error

Other Parts Discussed in Thread: AM3359, SYSBIOS

HI

I'm testing AM3359 ICE with full feature ethercat  appl. But there is weird problem..

  

when I send Safe-OP command from Pre-OP or Init (In Box 1 Online Menu), 'Invalid mailbox configuration' error was appeared and disappear-

-> and slave states is ....  Pre OP -> Init -> Pre OP -> Safe OP

[ error was that SM1 Enable bit of ' SyncManager1 activate reg (0x80e)' is disabled..]  What makes this bit disable???

   

BUT !!  when I send Safe-OP command from Pre-OP or Init (In Device 3 Online Menu), State Machine was changed to Safe OP without error

What is difference between two command???

below is my twinCAT Capture.. 

and my Test environment is..

board : AM3359 ice

Target os: sys/bios 6.33.05.46

AM335x sysbios industrial sdk 1.0.0.5 etherCAT application

EtherCAT slave Stack code 5.01

  • Hi Moonseon,

    I can't reproduce the error message here. Can you check the pre-compiled binary please and see if it has the same issue?

    If not this is a problem with your patch process or build steps. If you see the same with our binary please send us the Twincat project files so that we can try to reproduce.

    Are you sure you have installed the latest and matching XML device files for the slave? Such an issue should also show up during verification with the EtherCAT CTT and we have no errors using this tool. Are you able to run CTT on your setup?

    Regards.

  • Dear Moonseon,

    you can find a zip with pre-built binaries for the IA-SDK on the download page. There is a binary for the Ethercat example (lib version) only. This is our reference code. We are not able to test your changes on the application side.

    Regards.

  • Hello Frank

    As you told me. I tried to use pre-built binary in this morning. result was successful.. No error during 'State Machine is changing '

    I think this binary same as "not full feature EtherCAT example" .. is this right??

     

    and second question is...

    How did you makes this binary??

    In my case.. 

    SDK(1.0.0.5) and SSC tool(1.1.1.0)  are used... and SSC (5.01)..  of course setting is "TI AM335x Sample" and

    I had to add " #define TIESC_HW 1" in ecat_def.h 

    ESI file is used one in "c:\ti\am335x_sysbios_ind_sdk_1.0.0.5\sdk\protocols\ethercat_slave\ecat_appl\esi\"

     

    Please tell me about what makes difference..

    Sincerely 

  • Hi Moonseon,

    you may use the SSC tool but then you need to ask the Beckhoff/ETG team for support. Alternatively you can use the patch process as described in the IA-SDK documentation. That should yield to sources that compile out of the box and are identical with our versions. No need to set some defines as this is done by the patch.

    Regards.

  • Hi again Frank

    I'm still stuck in this problem .. please help me..

    few hours ago..  I tested example that is made from SSC5.01 (Patched TI SDK)... and copy all SSC files to example folder .. "Not using SSC Tool"

    But still same problem happens..

    I just need what you have same as "Library example"... 

    and here is my Wire_shark capture file ..

    and if you want see my problem in your PC

    you should change EtherCAT state Machine in EtherCAT slave Box Not in Device (Master) Menu...  (see picture)

  • Hi Moonseon,

    I am sorry you still have issues with your development. But I think it is not due to our code, right? Instead you are trying to build your own application with a different device configuration and description, correct? In this case of course you may have errors or mismatches in your code that you need to debug. The error messages from TwinCAT can help to know where to look at. Please do not post any SSC sources here as this would violate the Beckhoff/ETG rules.

    You may get additional support from Beckhoff/ETG on the stack itself. Also the EtherCAT forum is usualy quite responsive on such matters. As a last resort I can recommend the workshop trainings available from ETG that explain the EtherCAT slave process in full detail.

    Best regards.

  • Hi,

    This issue is seen when state transition (PREOP to SAFEOP) is performed at slave online tab and LRW support for process data is enabled. We have identified the root cause and fixed this issue in the latest firmware which will be part of 1.0.0.6 release (Mid December time frame). With 1.0.0.5 release, work around is to disabled LRW support for process data (Slave selection > EtherCAT > Advanced settings > Behavior > Check Use LRD/LWR instead of LRW ).