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.

TMDXICE3359's maximum EtherCAT SYNC0 frequency

Other Parts Discussed in Thread: AM3359

Hi,

I want to use EtherCAT DC SYNC0 frequency(minimum 2kHz).

But when I try to decrease SYNC0 cycle time under 4ms(250Hz) in TwinCAT, EtherCAT Statemachine is in "ERR SAFEOP".
(running TI example EtherCAT project..) 

So, I want to know maximum SYNC0 frequency in use of EtherCAT communication.

Did you ever test this?

  • We have tested until a cycle time of 250 ms (frequency 4 kHz) with a TwinCAT - Intel PCI Ethernet Adapter (Gigabit) on a PC / TwinCAT master and up to 100 ms (10 kHz) using a PLC master.

    Could you tell us the AL Status Code (register 0x134) reading? If it reads 0x0032 (PLL Error), the problem should be resolved using a compatible network adapter.

  • If you make the numbers 250us and 100us it will be correct...

    Yes, we used 10kHz DC sync in real application demos. You can find that on videos on the internet.

    Regards.

  • Frank,  do you have a link for these videos?  I am just trying to learn how to implement DC on the AM3359, and am pretty confused. :)  Thanks!

  • Yes, I'm suffered that PLL Error.

    But I'm using "Intel 82574" network adapter and this is listed in this page..
    (http://infosys.beckhoff.com/english.php?content=content/1033/tcsystemmanager/reference/ethercat/html/ethercat_supnetworkcontroller.htm)

    Can you tell me your master pc specification? (cpu, mainboard, etc...)

    I tried to buy beckhoff IPC, but that is too expensive....

  • Did you tested in bechkoff IPC?

    Can you tell me your test environment??
    (PC Specification...)

    I think there's not only network adapter but also invisible something effect EtherCAT communication.
    But I don't know what it is.... 

  • PeterN said:

    Frank,  do you have a link for these videos?  I am just trying to learn how to implement DC on the AM3359, and am pretty confused. :)  Thanks!

    Here is a demo video : http://www.youtube.com/watch?v=xw0bfjsk7Og

    But that is not for learning how to implement DC mode. I suggest you look at ETG pages and their trainings. Also the forum there should be helpful.

    regards.

  • taeyun kwak said:

    Did you tested in bechkoff IPC?

    Can you tell me your test environment??
    (PC Specification...)

    I think there's not only network adapter but also invisible something effect EtherCAT communication.
    But I don't know what it is.... 

    Yes, we tested against Cx5020 systems. I suggest you ask Beckhoff for DC requirements of TwinCAT, They also provide example applications that you can adapt for ICE board.

    Usually you need to run TwinCAT in Run Mode to get to DC OP state. This will require a PLC app or as a minimum a dummy task in the System Manager that is linked to one PDO value in your device. Also you will need two devices for real Sync op as the first slave is always just the clock reference.

    Regards.

  • Usually you need to run TwinCAT in Run Mode to get to DC OP state

    The steps for that are described here.

  • Thank you for your help, but this isn't work..

    I have problem like above image.

    EtherCAT State machine is in PREOP....

    I have followed step your document and used your .tsm file too, but results were same.

  • Taeyun,

    I have never seen this kind of errors (box 1001 and 1002). The first one is strange: "Frame missed 10 times" usually means there is no communication at all. Secondly the last one on Box 1 seems to indicate a wrong configuration as the device says it does not support DC. Obviously then you can't get to OP in DC mode config. But that should not happen as we have DC enabled in our XML files.

    Now I suggest you try the test using ICE and the precompiled binaries from the latest IA-SDK package (extra download) and the default XML. This should work if you follow the instructions and verifies that your test setup and procedure is ok. Maybe something went wrong when you build your binary or you have a mix up of configuration files.

    Regards.