I have a set-up where we have 7 VFD drives communicating through Modbus to a TCP/IP-Serial Gateway that is integrated with Allen Bradley Controllogix.
All 7 drives are wired in series as nodes 1 through 7. Communication is perfect between the first 5 drives - transmit and receive functions are working just fine. The moment I try to communicate to either node 6 or node 7 there is communication error and I lose all receive data from all other other drives, and transmit data becomes very delayed.
This sounds like a serial communications problem (wiring problem, baud mismatch, etc.). When the commands are sent to node 6 and 7, there is probably no response (or a bad response), so retries are sent out and fail and the device goes into error. Check that the wiring is correct, that there are terminating resistors at both ends of the bus, and the serial settings match between all devices.
I have the 480Ohm and 120Ohm resistors connected on the gateway side as per the manual, and a 120Ohm resistor at the final node.
What is interesting is that the drives at Nodes 6 and 7 get my transmitted data. They receive my run command and speed refs, but the error is still present in the function 16 write transmission and I get nothing on the receive (function 3) transmission.
When I am sending data to nodes 6 and 7, the Anybus will also periodically restart itself (all lights go out) and drop all communication until it starts back up. That seems unusual.
A few other observations:
When drives 1 to 5 are running, status light is solid green - no loss in comms. Transmit and receive are perfect.
I changed baud from 9600 to 57600 and now I can bring drive 6 online but status light starts to flash red a bit - Typically 3-4 green flashes, a bit of steady green, then some red, then back to green flashes as it repeats.
Bringing drive 7 will cause status light to go red and communication seems to crash.
Could there be an issue with the amount of data being transferred? My message delimeter is currently set to 0 (I have tried to adjust but ‘0’ seems to work the best)
That’s interesting that’s where it is, given the issue you’re having is with nodes 6 and 7, but maybe it’s just the last transaction sent. In any case, the message delimiter of 0 (1 second) should be sufficient to prevent an error.
Yes, after seeing the log file, I’m sure it’s not a wiring problem.
The log file doesn’t show any errors or delays. It looks like all 7 nodes are responding perfectly. I would like you to capture some more logs however, as it only captures about 1 round of commands. Maybe we are missing something that happens sporadically.
I think that should only happen if Module Reset is enabled:
I’m not 100% sure on that though and I’d have to check with the product team. Are you experiencing this?
Thanks for all that info.
It was the Subnet Status light (#5) that would periodically turn red and flash, most prominently when nodes 6 and 7 came online.
The restart is set as ‘disabled’ so not sure what else would cause the module to restart. It only occured when after active signals were transmitted to nodes 6 and 7.
However, all that being said, I have rewired nodes 6 and 7 and brought them to the start of the chain instead of at the end of the chain and that seems to have completely solved all the problems. No retransmission errors, no red lights for subnet status, and the network seems to be responding as quickly as expected.
I will continue to test and monitor, but it seems like everything is working as it should for now.
