Andreas Jacob
Hilscher Gesellschaft fuer Systemautomation mbH
Profibus Master-Slave Class2 Connection
Hello KG,
I am sorry, but this functionality is not included in the used library of the PROFIBUS Slave Example from the Hilscher web page.
The library of the provided example is limited in the functionalities.
That means you can not implement the requested "Master Slave acylic Class 2 Connection" functionality.
If you need the complete range of functions you have to buy the unlimited PROFIBUS Slave stack from Hilscher.
Regards
AJ
Assuming to have PROFIBUS Slave (not only the example on your side), is it principally possible to have on the system:
1xProfibus SLAVE and 1xProfibus Master Class2 (without Master class1)?
Is there a way to enable communication between the 2 devices in order to be able that Profibus slave intergration is up and running (of course without APP layer)?
Andreas Jacob
Hilscher Gesellschaft fuer Systemautomation mbH
At the moment we have a Profibus Master only with Class 1 and Class 2 services. We did not test a standalone class 2 master.
Of course it is possible to run a Profibus Master and a Profibus Slave at 1 netX chip. They have only configured to 2 different XC. And the 2 devices should also be able to communicate together.
Hey AJ
What we actually wanted to try is to connect a netX Slave to a external hardware that is a master class 2, not run both on the netX.
What we now tried is to connect a master class 1, the master class 2 for diagnostics and one netX slave. The problem is now, that the slave is somehow stopping.
Here you see what is happening on the communication level. Address one is our master class 1, address 32 is the netX slave with the demo stack.
Do you know why this happens? As you can see on the info panel on the left, the telegram with the DL service states (which I think is sent by the netX) states that Station_Not_Ready...
Andreas Jacob
Hilscher Gesellschaft fuer Systemautomation mbH
Hello HP,
the demo stack (from the Hilscher Web page) is limited and does not support DPV1 Claas2 services.
Still, it should be able to correctly communicate to the master class 1, right? And that it clearly did not do, because it was always in the stop state.
You can also see that this mentioned message above was sent from address 32 to address 1, which means it was communicating to my master class 1, not the master class 2, who is in this setup basically a telegram sniffer...
Andreas Jacob
Hilscher Gesellschaft fuer Systemautomation mbH
The profibus slave is at the state wait parameter. The reason for that state is not shown at the bus trace. Can you please give us your Set Cfg Telegram and Set Prm Telegram for better diagnostic.
The Demo Stack only support the Identnumber 0xa12 and the cfg bytes 0x21 0x11.
I'm out of office today. I'll try to capture that on Monday. However, I think this is the complete communication that was going on between the master and the slave...
I switched to the NXEB 100-NET and a complete new profibus application on my master, to make sure it's not a problem of my own target or the old application I copied. Still, the slave is not going into data exchange mode. Here's a trace of the startup communication:
As you can see, there is no Set Cfg message, only a Set Prm. Is that the problem? If yes, how can I provoke such a message from my master?
It seems to be a problem of my master class 1. I was using a PCI-Card from Comsoft and a LabView application. That worked nicely for other applications. For this one it did not. It works when I use an S7 SPS though. Slave is now going to data exchange mode.
Hi HP,
I think the slave runs into watchdog and request new parameter data. Your PCI Card shall also run. Maybe you can set watchdog time to a bigger value. At my example I have 200 ms.
Thanks for the feedback, I'll try that...
Andreas Jacob
Hilscher Gesellschaft fuer Systemautomation mbH
Hi,
no there is no manual with further information reagding the error codes available.
But "TLR_E_RCX_RX_KNL_TOKEN_INVALID" means, that the used Kernel Token is not valid.