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.

AM2432: Cable Redundancy Loopback Feature Link Break

Part Number: AM2432

Hi TI Experts,

Customer has SOP AM2432. They are going to add the cable redundancy topology loopback feature soon, you could refer to the following details.

https://infosys.beckhoff.com/english.php?content=../content/1033/ethercatsystem/2474143371.html&id=

 

However, customer found that previously there was another customer below, who incorrectly connect in a reverse way and having DC sync about 680ns delay. My customer thinks that, although this situation should be avoid, if they later on implement the redundancy topology loopback feature, and if the link breaks, the rest part will act 

as revert insertion state, which could be in the same situation as the followinng thread. And in the following thread, you could see that with just two axes, the Sync0 offset already reach to about 1µs, nearly surpassing the <1µs synchronization claimed by EtherCAT DC. Customer has a great concern on it.

 

Hence, could you help test the standard cable redundancy topology on your end and measure if there is any obvious delay if there is link break happens to address customer's concern.

 

Customer is going to use the MainDevice Port 0 & Port 1.

 

LP-AM243: EtherCAT: DC sync has a 680ns delay - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

 

Thanks,

 

Kevin


  • Hi,


    Our expert has enabled Redundancy mode in the following topology and tested DC synchronization:

     

    • TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> AM243x (D1) <> AM243x (D2) <> AM243x (D3) <> CU2508 (Port X2)

     

    Reference Node or DC Synchronization: CU2508
    Cycle time tested: 1ms
    Firmware version: 6.5.72 (x0548)
    Sync probed between D1 and D3

     

    Initial Analysis:

     

    Scenario

    Device Type

    Network State

    Sync0 Offset

    (D1 vs D3)

    Jitter

    Normal Operation

    TI AM243x (D3)

    DC + Redundancy Enabled

    +20ns (right shift)

    15-20ns

    Link Break (D2 – D3)

    TI AM243x (D3)

    OP with LINK_MIS A/B

    -120ns (left shift)

    15-20ns

    INIT→OP with Link Break

    TI AM243x (D3)

    Starting with broken link

    ~750ns (left shift)

    15-20ns

    Normal Operation

    EL9800 (as D3)

    DC + Redundancy Enabled

    +90ns (right shift)

    15-20ns

    Link Break

    EL9800 (as D3)

    OP with LINK_MIS A/B

    -50ns (left shift)

    15-20ns

     

    This is the data our expert has recorded with the above mentioned setup. EL9800 is a Beckhoff Reference Board which is used to verify the EtherCAT SubDevice implementation.

    Please help review if there is anything further need our expert to check.

    Thanks,

    Kevin

  • Hi Kevin,

    So I've run couple of tests with more varied topologies with TI ESC and EL9800 as follows:

    Environment Configuration:

    • EtherCAT MainDevice                                : TwinCAT3
    • PLC                                                             : C6015-0020
    • Reference Node or DC Synchronization    : CU2508
    • Cycle time tested                                       : 1ms
    • Firmware version                                       : 6.5.72 (x0548)
    • Sync probed between D1 and D3

    Topologies tested:

    1. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> AM243x (D1) <>  AM243x (D2)  <> AM243x (D3)  <> CU2508 (Port X2)
    2. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> AM243x (D1) <>  AM243x (D2)  <>  EL9800 (D3)  <> CU2508 (Port X2)
    3. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> EL9800 (D1)  <>  AM243x (D2)  <>  EL9800 (D3)  <> CU2508 (Port X2)
    4. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> EL9800 (D1)  <>  EK1100 (D2)  <>  EL9800 (D3)   <> CU2508 (Port X2)
    5. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> AM243x (D1) <>  EK1100 (D2)  <>  AM243x (D3)  <> CU2508 (Port X2)
    6. TwinCAT (C6015 PLC) <> CU2508 (Port X1) <> AM243x (D1) <>  EL9800 (D3)  <>  AM243x (D3)  <> CU2508 (Port X2)

    Scenario

    Topology

    Network State

    Sync0 Offset

    (D3 vs D1)

    Jitter

    Normal Operation

    1

    DC + Redundancy Enabled

    +20ns

    (right shift)

    15-20ns

    Link Break (D2 – D3)

    1

    OP with LINK_MIS A/B

    -120ns

    (left shift)

    15-20ns

    INIT→OP with Link Break

    (10 iterations)

    1

    Starting with broken link

    ~750ns

    (left shift)

    15-20ns

    Normal Operation

    2

    DC + Redundancy Enabled

    +90ns

    (right shift)

    15-20ns

    Link Break

    2

    OP with LINK_MIS A/B

    -50ns

    (left shift)

    15-20ns

    INIT→OP with Link Break

    (10 iterations)

    2

    Starting with broken link

    ~750ns

    (left shift)

    15-20ns

    Normal Operation

    3

    DC + Redundancy Enabled

    0ns

    15-20ns

    Link Break

    3

    OP with LINK_MIS A/B

    -140ns

    (left shift)

    20ns

    INIT→OP with Link Break

    (10 iterations)

    3

    Starting with broken link

    ~720ns

    (left shift)

    20ns

    Normal Operation

    4

    DC + Redundancy Enabled

    0ns

    20ns

    Link Break

    4

    OP with LINK_MIS A/B

    -50ns

    (left shift)

    20ns

    INIT→OP with Link Break

    (10 iterations)

    4

    Starting with broken link

    -1.22 μs

    (left shift)

    20ns

    Normal Operation

    5

    DC + Redundancy Enabled

    0ns

    20ns

    Link Break

    5

    OP with LINK_MIS A/B

    -60ns

    (left shift)

    20ns

    INIT→OP with Link Break

    (10 iterations)

    5

    Starting with broken link

    -1.24 μs

    (left shift)

    20ns

    Normal Operation

    6

    DC + Redundancy Enabled

    0ns 20ns

    Link Break

    6

    OP with LINK_MIS A/B

    -80ns

    (left shift)

    15ns

    INIT→OP with Link Break

    (10 iterations)

    6

    Starting with broken link

    ~700ns

    (left shift)

    15ns

    Do note that the jitter is measured after the signal stabilizes (that is, not a long term measurement).

    Reference:

    Regards,
    Aaron

  • To summarize the above table, the TI ESC behavior is similar to ET1100 for DC synchronization in Redundancy mode.

    Regards,
    Aaron