Essential Reading: 30 Proven EtherCAT Diagnostic Tips for Beckhoff Systems

完成条件

1. Analysis Method for EtherCAT CRC Errors

Refer to this diagram, how can we determine which module is most likely the source of the issue, and what is the basis for this determination?

TX/RX error refers to errors detected at the physical layer, encompassing issues with the network port section or the EBUS portion. These errors occur ten times in the queue section, where the queue holds aperiodic data frames. Modules that contain a mailbox section are the EL2622, EL3062, EL3044, and EL5151.

Lost frames indicate the loss of data frames, with 14 periodic data frames missing in the status. This can be challenging to pinpoint as modules such as the EL2622, EL3062, EL3044, and EL5151, which incorporate mailbox data, also include process data and can experience the loss of process data frames.

The CRC (Cyclic Redundancy Check) is processed in the automatic switching portion of the slave communication chip's port, hence CRC detection takes place in all ports utilized by the slave. This process does not occur within the data frame processing unit. The presence of CRC in an NO30 module signifies an error in a data frame. The corresponding port CRC verification sequence is illustrated in the following diagram:

The CRC verification counter for the EL3062 module increases the most, indicating a higher likelihood of issues occurring at the inlet of the EL3062.

2. Function and Role of Synchronization Unit Grouping (Sync)

To prevent abnormal behavior of certain slaves from interfering with other devices or to facilitate fault source identification, the synchronization unit data grouping feature can be employed:

Many Beckhoff customers may be unaware of the function and purpose of setting up synchronization units in EtherCAT communication.

A synchronization unit, in the most straightforward sense, divides the main data frame of EtherCAT into sub-data frames corresponding to each synchronization unit.

For example, if there are 2 EL1008 modules and 1 EL3204 module, and they are defaulted to operate under a single synchronization unit (Sync), these three modules share a single sub-data frame. If one of the modules encounters a problem, the data of the other two modules cannot be refreshed. However, if two synchronization units are configured, Sync1 (for 2*EL1008) and Sync2 (for EL3204), the main data frame of EtherCAT is divided into two corresponding sub-data frames. The sub-data frames of the synchronization units do not interfere with each other. If an EL1008 encounters a problem, causing a CRC verification exception in its corresponding data frame, the other EL3204 can still function normally and its data is refreshed as usual.

It is highly recommended that when creating TSM files, multiple synchronization units should be configured.

Grouping Method: Assign a unique Sync Unit Name to units that require independent operation. Units with identical names belong to the same data group.

Strongly recommended modules to be configured as separate synchronization units:

• Safety PLCs: EL6900, EL6910, ...
• Safety slaves: EL1904, EL2904, EP1908, ...
• Fieldbus gateways: EL6731, EL6751, EL6652, ...

• When scanning independently is feasible with OP, but anomalies occur when used alongside products from other manufacturers: Third-party drives

3. Emergency -> Scan Module Scanning Function
   In the TwinCAT System Manager, there is a utility that can scan EL modules for initial diagnosis.

4. First, connect to the Beckhoff PLC controller and switch to Config or Free Run mode.

5. Navigate to EtherCAT -> Advanced settings -> Emergency -> Scan to scan the modules. Refer to the diagram below for illustration.

4.Consequences of Cable Interruption in EtherCAT Network
The output of all subsequent modules after the interruption point will become zero.

5.Are there any effective methods to quickly identify the faulty module?
Apart from physically unplugging each module one by one to locate the damaged one, there are two additional methods that can assist users in determining which module is malfunctioning:

1、Examine the CRC and Changes counters within the EtherCAT -> Online section. If the CRC of a specific module displays a non-zero value, it signifies that there is an issue with the modules preceding, following, or with the module itself.

2.For CE systems, it is necessary to first access the controller's control panel and enable the error logging function:

By default, the save path for the error log is set to "My Computer" and the file is named "tcsyslog.txt":

If you are using Windows 7/10, you can click on the TwinCAT icon in the bottom-right corner, navigate to Tools –> EventViewer to check if there are any error logs recorded.

6.What should I do if the embedded controller is unable to scan subsequent modules?

It could be a fault with the E-BUS or a poor contact between modules. In case of poor contact, you can try reseating the module or firmly pressing down on both sides (this operation should be performed while the power is off).

Alternatively, you can connect to the EK1100 via the controller's Ethernet port and use the EK1100 to test the modules. If the scan is successful, it is likely that the fault lies with the E-BUS of the controller itself.

6.What is the issue when setting a digital output to true but the output keeps flickering?
The cycle time of the Task is set too long. It is recommended to set it to less than 100ms.

7.When running the online controller, a popup error 1861 appears;

during this time, executing a scan results in error 179

8.Runtime error: The slave displays "received invalid DC timings! check device state for INIT, communication error '1823'".

After rescanning, the fault code 1798 appears: It is inferred that there may be a fault with the slave or the IN and OUT ports of the slave are incorrectly reversed. The only way to troubleshoot is to disconnect the Ethernet cables from some of the slaves and perform a scan. Eventually, the faulty slave was identified and replaced, resolving the issue.

9. In TwinCAT 3, to add the Changes counter for observation in the EtherCAT network under IO in the Online tab, simply select the EtherCAT network from the Online tab and then choose Properties.

In the pop-up dialog box, tick the box next to "Show Change Counters (State Changes/Not Present)" on the right side, and then click OK to confirm.

However, in version 4024.7, the aforementioned options will not be available when right-clicking (as shown in the figure below).

At this point, we can access the desired settings by selecting the EtherCAT network, navigating to the EtherCAT tab, and clicking on the "Advanced Settings..." button.

In the opened dialog box, select "Diagnosis-Online View" and tick the checkbox next to "Show Change Counters (State Changes/Not Present)" on the right side. Then click OK to confirm.

10. What does it mean when some of the "repeat sup" entries in "sync unit assignment" have a cross (X) mark?

Question: What does the cross (X) mark in "repeat sup" under "sync unit assignment" indicate?

Answer: The X indicates a digital module.

11.How to automatically calculate E-bus current consumption?
To calculate E-bus current consumption, first configure the modules according to the system requirements during the configuration process. Then, in Device X, under the EtherCAT tab, you can find the E-bus current consumption statistics as shown in the figure below. This allows for convenient access to information on E-bus current usage.

When the current of the E-bus is insufficient, it is necessary to add an E-bus power terminal module, such as EL9400, EL9410, etc.

12. Methods to Identify Modules with State Not Equal to OP

Method 1:
Create an array where each element is linked to the State value of individual modules. Check if the State value is not 8 (OP) to identify modules not in OP state.

Speed of Information Acquisition: Fast within the current cycle.
Suitable for: Fixed configurations to locate problem points.
Limitations: Cannot simultaneously obtain information such as disconnection status.

Method 2:
Function Block FB_EcGetAllSlaveStates,

How long does it take to obtain values?
The PLC invokes this function block and obtains data within two Task cycles, with minimal impact on the CPU.

What values are obtained?
It can retrieve deviceState and linkState information.

DeviceState: INIT, PreOP, BOOTSTRAP, SAFEOP, OP

13.Summary of SetSlaveState

(1) FB_EcSetSlaveState can be used to individually control the restart of an EtherCAT slave. When the PLC utilizes this module to issue commands to the EtherCAT master, the master integrates these commands and sends them to the slave, potentially causing an increase in CPU utilization.

(2) Different modules take varying amounts of time to transition from INIT to OP. Digital modules typically have a shorter transition time, while analog modules take longer, and communication modules such as EL6731 and EL6751 may require approximately 100 cycles.

(3) The performance of the controller CPU and the number of modules set to OP can have varying impacts on the CPU. For instance:

• CX1020, with a 1ms Task and 16 slaves, running SetAllToOp can instantaneously reach over 9692 (96%) utilization, with a full-scale value of 10000. Running SetOne, on the other hand, can reach around 1300 (13%) utilization.
• CX5020, with a 1ms Task and 30 slaves, running SetAllToOp can exceed 15000 (150%) utilization, despite a full-scale value of 10000

14.Impact on Other Modules within a Synchronization Unit Group When WcState of a Unit is 1

When the WcState value of a synchronization unit equals 1:

• DI/AI values remain unchanged from the previous cycle.
• DO/AO values depend on the State of the module itself:
• If the module's State is 8, it can output normally according to the PLC program.
• If the module's State is not 8, after the watchdog timer expires, the module's output is set to 0.

15.How IO is Refreshed in EL Modules During SafeOp, PreOp, and INIT

During SafeOp:

• Input values are refreshed periodically.
• For digital outputs, there is no output after the set watchdog time expires.
• Analog outputs are immediately set to 0, but the actual output voltage reaches 0 within 20ms (turn-off time).

During PreOp and INIT:

• AI is immediately stopped from refreshing.
• AO channels are immediately set to output 0, with the actual voltage reaching 0 within 20ms (turn-off time).

16.EL Module Status Alternately Flashing Between SafeOp, init_err, OP, and INIT
The fault symptom arises when a variable mapped to a module is allocated to a large task instance (100ms). Deleting the mapped variable restores normal operation for all modules.

17. The EL module status alternates between flashing "safeop init_err," "op," and "init."
    The fault phenomenon arises due to a variable mapped to a module being allocated to an instance of a significantly large task (100ms). Upon removing the mapped variable, all modules return to normal operation.

18.     Error Reported During EtherCAT Module Switching to OP: "AL status '0x0012' read and '0x0004' expected. AL Status Code '0x001e - Invalid SM IN cfg'"

Taking the EtherCAT bridge module EL6695 as an example, after the master station configuration is completed, the PDO (Process Data Object) scanned from the slave station does not match that of the master station

When adding input and output variables to the Prim end of EL6695, selecting Bool is not allowed, as it would cause the El6695Box scanned by the Secondary end to read Bool variables as bits, resulting in inconsistent PDO data sizes and preventing the Secondary end from entering OP mode.

Solution 1: Add Byte variables instead of Bool variables and bind them to the Bool variables in the PLC. (This is because both Byte and Bool occupy 8 bits in TwinCAT3.)

Solution 2: Add Bit variables instead of Bool variables and bind them to the Bool variables in the PLC. (This has been tested and found to be feasible, however, it is still recommended to use Solution 1 with Byte variables.)

19.     When scanning I/O with the EK0000 module, the module fails to be recognized properly, and an activation configuration error 1798 (0x706) occurs, accompanied by "IoOpen failed" and the module status indicates "int_err". A resolution method involves manually writing to the EEPROM.

The fault phenomenon occurs during the scan io process, where the device name fails to display correctly (despite the fact that the actual

Solution: Attempt to manually write the EERPOM on the abnormal slave station using the original configuration, then rescan the module. If it enters the OP state, it indicates success. If the write operation prompts an operation not allowed or the error persists as "int_err", please contact Beckhoff personnel for further assistance.

20.How to maintain output state when an EtherCAT module experiences communication abnormalities?

In the slave's configuration page, navigate to: ethercat → advance settings → behavior → watchdog → set sm watchdog. Check the box and set the value to 0.

Alternatively, in the Behavior section under General, locate WatchDog:
Check the box for Set SM WatchDog and set the value to 0.

21.What is the meaning of AL Status Code?
The AL Status Code is diagnostic information at the software level related to EtherCAT state operations. Once a slave fails to enter the correct state as required by the master, an AL status code will be reported in the slave's register word.

22.In the Topology View interface, what do the small circles under each module represent?
The small circles under each module in the Topology View interface indicate the presence of CRC (Cyclic Redundancy Check) errors on the module's ports A, B, C, or D.

23.What does ERR SAFEOP indicate?
The presence of "SAFEOP to OP failed!" in the software error list indicates a state-related issue. Verify the EtherCAT slave state by accessing the online status through the device's (EtherCAT) online interface.

1. If it appears in the status bar of the EL module, please first confirm that the E-bus power supply is sufficient. If power insufficiency is detected, please add power supply modules to the insufficient locations.

2.      If this error appears on all modules, please verify the controller's current license to ensure it includes the necessary IO control level, or check if the license expiration date has passed (a discharged controller battery can reset the system time, requiring battery replacement).

3.      For an AX5000 displaying "safeop err preop init_err", check the safety board settings and safety signal status in the servo's DriveManager under the Safety Option. If signals are missing, trace back to the safety logic for confirmation. (For non-servo devices encountering "preop init_err", it's typically due to damaged Ethernet cables or external hardware anomalies. Hardware additions/subtractions may be necessary for troubleshooting.)

4.      If modules were previously working normally but now multiple modules display "safeop" accompanied by other errors such as "init" or "mis_lnk", scan the configuration to confirm proper power-up and wiring of the actual hardware.

5.      (Duplicate of previous point, but with additional information) If modules were functioning normally but now display "safeop" under a new configuration, specifically for high-function modules, attempt to initialize the module by navigating to the module's CoE Online and entering 16#64616F6C into 1100:01.

6.      If modules were functioning normally before, and the issue resolves after a restart but recurs occasionally, or if step 4 does not resolve the issue, it is recommended to replace the modules with spare parts.

7.      If the system can achieve OP status in free-run mode but encounters errors upon activation, the issue is likely related to configuration. For third-party slaves, please verify their status and ensure proper configuration (contact the manufacturer for description files and configuration documents compatible with the current software version). If the third-party product functions normally when used independently, focus on the integration and configuration aspects.

24.How to Configure the WcState Fault Tolerance Period in TwinCAT 3？

In actual field applications, issues often arise due to invalid data (WcState=1), such as NC error 4466, which can sometimes be challenging to resolve on-site. To address such problems, one approach is to appropriately increase the WcState fault tolerance period. It's important to note that invalid data issues are typically caused by communication issues, and this method does not address the communication problem itself.

For instance, when encountering NC error 0x4466 (17510),

Upon reviewing the help documentation, we have discovered that error 4466 is actually triggered by three consecutive NC cycles with invalid data (WcState=1).

The field input module is in an OP (Operational) state with the input light illuminated, yet there is no input registered in the program. This is also attributed to invalid data (WcState=1).

Method for Configuring TwinCAT 3 WcState Tolerance Cycles

To adjust the frequency of invalid WcState data, navigate to the corresponding Sync Unit and tick the box labeled "Individual WcState Tolerance Cycles" (this setting takes effect after activation of the configuration). Through this option, you can set the WcState Tolerance Cycles. For instance, if you set WcState Tolerance Cycles to 2, it means that only when two consecutive cycles have invalid WcState data will a value of 1 be written. The maximum number of cycles that can be set is 16. Different Sync Units can have their individual WcState Tolerance Cycles configured.

     

   Note: Some third-party servos or slaves have corresponding settings within their internal configurations. If TwinCAT3 is configured with this setting, the corresponding fault tolerance values of the third-party drives should also be increased accordingly.

25.The routing keeps disconnecting and then automatically reconnecting. How should I handle this issue?

1．Loose Ethernet cable connector or there is interference.

2.Communication timeout.

26.EtherCAT Communication Instability
This issue is typically resolved by using an Intel network card that is supported on the compatibility list.
If the network card is not on the supported list, there is a risk of encountering a blue screen error during hardware scanning or when activating the configuration. If the blue screen is avoided, communication errors may occur frequently.
I have tested two computers that previously experienced blue screens, and after reinstalling the operating system (with drivers installed), installing TwinCAT 2232 ran without issues. Currently, I am running EtherCAT on my laptop with 50us tasks without any problems.
It is recommended to use a clean operating system with all necessary patches and updates applied.

 

27.What do the slave statuses "OP Link_Mis A" and "OP Link_Add B" displayed by the TwinCAT master station mean?
It suggests a wiring error on your side. The A port, which is supposed to be connected to an Ethernet cable, is not actually connected, while the B port, which does not require a connection, is mistakenly wired.

28.The E-bus voltage of the first module after the coupler displays a negative value.
When using an EL module directly connected to a CX embedded controller, the power supply wiring and voltage measurements are all normal, but the first module displays a negative E-bus voltage as shown in the image below. At this time, the safety module cannot be added to the Safety project.

It was discovered that the coupler module is not being displayed, whereas the CX embedded system's built-in coupler should be shown as EK1200, but it is missing.

At this point, the best approach would be to check if any EtherCAT description files, particularly those that come with TwinCAT, have been deleted from the debugging PC. Update the description files to ensure that all built-in EtherCAT description files are present. After doing so, reopen the software and scan for IO, which should then function normally.

The storage paths for TwinCAT3 description files are:
C:\TwinCAT\3.1\Config\Io\EtherCAT

And for TwinCAT2 description files:
C:\TwinCAT\Io\EtherCAT

29.What functionality can be used to diagnose when the actual task cycle time exceeds the preset time?
The watchdog function of the bus terminal module can be configured to detect instances where the actual cycle time exceeds the preset time. If the module fails to receive valid process data from the controller within the preset time, the watchdog will trigger an alarm after 100ms.
The watchdog function can be activated using KS2000 or through TwinCAT. For Profibus, this function can be set up on the Profibus master station.

30.What causes the EtherCAT state machine to periodically transition from NO_COMM to OP every 5-10 minutes and then recover?

Possible Cause: The use of single-core wires, where alternating friction can generate CRC errors, and slow detachment may lead to a probability of losing 10 frames. (Shortening the scan cycle has minimal impact.)

Solution: A faulty connector in a certain section of the Ethernet cable resulted in the EhertCAT master station continuously losing 10 frames and restarting. The issue was resolved by replacing the Ethernet cable.

完成