Can Open Assistance Please

Greetings, I would like to hear from the members of this group who have a wide knowlege of CanOpen. I know that CanOpen is much faster than say something like Modbus.

What I am tryng to do right now is have a Horner XLT talk via CanOpen to a Lexium 05 Drive. I have followed just about everything I can think of and feel that the registers are configured correctly. However it seems as though I keep getting an NMT error and it seems to be with the slave, which is the Lexium Drive. if I take a twido PLC and using CanOpen connect to the lexium drive it communicates fine, but for some reason the XLT from Horner just wont seem to work properly. Was wondering if anyone on here has tried to integrate using CanOpen and what pitfalls and such I should look out for. Also it would be really helpful if someone on here had integrated a Horner to a Lexium using CanOpen.

I am wits end. Any and all assistance greatly appreciated. I have attached the program as I have it started here so as if you want you can look at my CanOpen config maybe someone here is more familiar with CanOpen can see where the communication error is. Thanks and have a great day,.

I have done a few canopen installations and I guess would consider myself "dangerous with it". But I have always used CoDeSys masters.

With Codesys, you read in the EDS file for the device and then map parameters into the PDOs. You can enable or disable the master doing a mapping in the slave a start up to write the proper object IDs to the proper places in the slave to map the PDOs.

Since you have one master that is working and one that is not I'm guessing it is something like this causing you the error. Other things to look for are the type of PDO transmission (cyclic, Synched, MFG specific), type of heartbeat or nodeguarding, although obvious, does the master have the baud rate set to match the slave, does the master have the proper termination, are CAN_H and CAN_L reversed on the failing master?

The good news is that you know the slave works.

I'm not really on front of any information right now bu these are the things that come to mind.

Norm, thank you for the posting. I am at this point still not further ahead than I was when I posted the question.

Here is what I have done so far to find out what the problem is. I have checked the cable using the master that works and the master that does not. The master that works is an M238 PLC, the master that does not is a Horner XLT. And before you ask NO I have only ever had one hooked up at any given time. Since the XLT and the M238 use the same type of port connection for the CanOpen it was easy to change it back and forth. And again before you ask YES there are termination resistors on the wire. So the wiring seems to be ok.

The next thing I have done is completely wiped the system and also done a refresh to the XLT firmware, just to make sure nothing got corrupt or anything. The XLT and Lexium 05 drive appear to be working fine it is just trying to get them to communicate via CanOpen.

Now using the Horner CSCAPE software I have gone in and configured the Master, I loaded the Horner EDS file and did not configure and SDO or PDO, I then added slave and again loaded teh EDS file for Lexium 05, and again there are no SDO or PDO configured.

Upon intialization for a period of about 10 seconds I get one error on the master status register, that error says that a slave has failed. And I get an error on my slave status register. This error message says:
Invalid NMT state (Mismatch between Master NMT state and that slave NMT state)
I am not 100% sure exactly what that means, does that mean that EDS can not communicate with the drive or does that mean the Lexium EDS can not communicate with the Horner EDS?

What kinds of things should I look at here to help diagnose and troubleshoot this issue? Any help greatly appreciated. Thank you and have a great day.

ndzied1 said:

I have done a few canopen installations and I guess would consider myself "dangerous with it". But I have always used CoDeSys masters.

With Codesys, you read in the EDS file for the device and then map parameters into the PDOs. You can enable or disable the master doing a mapping in the slave a start up to write the proper object IDs to the proper places in the slave to map the PDOs.

Since you have one master that is working and one that is not I'm guessing it is something like this causing you the error. Other things to look for are the type of PDO transmission (cyclic, Synched, MFG specific), type of heartbeat or nodeguarding, although obvious, does the master have the baud rate set to match the slave, does the master have the proper termination, are CAN_H and CAN_L reversed on the failing master?

The good news is that you know the slave works.

I'm not really on front of any information right now bu these are the things that come to mind.

Click to expand...

Since you are trying to get the Horner to Work, have you read this manual?

http://www.heapg.com/OcsManuals/CANopen/User Guide/CANopen User Guide.pdf

It mentions a special firmware for canopen units and loading EDS files and looks like a must read although i have only browsed it.

There also seem to be some quick start CanOpen Guides here:
http://www.heapg.com/OcsManuals/CANopen/Quick Start Guides/


You didn't mention if you checked the heartbeat between the two masters. That would definitely throw a monkey wrench into the mix.

I think because the Twido and Lexium are both Schneider products that with that combination there is some behind the scenes configuration that takes place you don't see. When you go to the Horner, then you will have to make all the configurations yourself. I won't have time to look at much more today but read the documents I pointed to and see if there is any light shed on the problem.

I'm not familiar with either the HMI or the drive that you list, but I'm familiar with CANopen.

"Invalid NMT state" may indicate that the HMI is expecting the drive to be in the operating (RUN) state, and the drive is actually in either the pre-operational or stopped state. There are a couple of conditions that can cause this. One reason may be that the drive has been setup to receive either node guarding or life guarding NMT messages from the PLC and the HMI, not being a master, is not sending these messages and the drive is dropping out of the operational state. It might also be the simple fact that the HMI is just not putting the drive into the operational state.

I can't suggest a guaranteed solution for you, but some things to try are:

1. Many CANopen devices have an option to boot-up and switch automatically to the operating mode without needing a master to initiate them. This is set in object dictionary value 1F80 bit 2. Check your drive documentation to see if this available.


2. Check the HMI documentation for a feature that will have it act either as a network master, or for a feature that will have it send the NMT message to switch the drive to the operating state.

3. Turn off any node guarding or life guarding features by setting object dictionary values 1017, 100C and 100D to zero.


Regards,
Frank