Troubleshooting EtherCat

See also chapter 9-5 in manual W505 and P114.

The image above shows how a data frame moves between the nodes. The yellow outgoing frame travels all the way to Pos 8 via Pos 6, which is a junction, before returning by the shortest route to the master.

The blue boxes update the part of the frame that concerns that node. So at Pos 8, all nodes have received their data and have also updated the telegram with the data to be sent back to the master.

All nodes have 4 internal ports, but only the junction (GX-JC06) has a physical connection to all four. Ports that are unused or disconnected are automatically bypassed.

Here you can see how well your nodes are communicating.

If the CRC count is increasing, you most likely have an electrical noise problem. Check the shielding and so on. It can honestly be hard to find – real ghostbuster work. It's better to reinforce and review the EMC/EMI countermeasures and the wiring layout (EtherCAT cables should not share cable ducts with noise sources).

Click Get to update the data. Each reading is saved with a timestamp that you can bring up from the menu at the top.

To export it, select Output File and a "Viewer"-file. That makes it easy for others helping with the troubleshooting to see the setup (via Tools → EtherCAT Diagnosis).

Information for each node is shown in the field at the bottom right after selecting a node on the left.

Firmware 1.64 added more function blocks for EtherCAT diagnostics. The data in the image above comes from EC_GetMasterStatistics and EC_GetSlaveStatistics, which give some of the same information as the diagnostics view. ErrCnt[] can then be monitored to give an early warning that something is wrong at a node, as described above. Use EC_Clear… to reset the counters.

In this example, node 16 has 252 errors on Port B. That is incoming data from node 18 (a junction), which has the same number of errors on X3. You need to work your way further down until you find the last node in the network with errors on its incoming port (Port B). The fault is then between that node and the next one below it – typically between 22 and 21 in the example.

• The number on Port A / IN / ECAT IN is the number of error frames coming from the node above.
• The number on Port B / X2 / X3 / OUT / ECAT OUT is the number of error frames returning from the node below.

Check the controller state code

Check Info 2

1 = Abort
2 = Input Refresh
3 = Timeout
255 = Other

Index:Subindex

Example:
Index = 1008
Subindex = 0021

Abort code