“PROFIBUS, PROFINET, and Fieldbuses in the news” in which Carl refers you to news of interest and makes a few comments.
Let’s start with the generic networking article from Control magazine: “Networking that floats all boats” which starts out by disparaging all fieldbuses before moving on to cover EDDL, OPC UA, Ethernet, and wireless. I have to take exception to the second paragraph’s “users still don’t have the completely open, interoperable, easy-to-use networks they need.” PROFINET is completely open, interoperable, and easy to use. Open as in having the standards specification available from an international body, the IEC. Interoperable as in using proxies to connect to other fieldbuses. Easy-to-use as in being able to configure that interoperability rather than having to write PLC code.
Take the opening of the next paragraph… please:
Selfish, proprietary interests continue to hold uninformed users hostage by subtly feeding their fears about new networking technologies, and then adding unnecessary layers of complexity to fieldbus protocols to confirm those fears. This traditionally is done to “protect users’ investments in their installed base,” which actually means “protecting the market share of suppliers too lazy to truly invest in innovation.”
No wonder some users are afraid to use a fieldbus. If the author is trying to single someone out, I wish he would so state. As a generalization this is just wrong. For example, with PROFINET we protect the competitive fieldbus’ installed base by using proxies to connect to their legacy fieldbuses. Why? Because it’s good for the users. It protects their existing investment. They can connect DeviceNet, Modbus, serial and other existing networks to PROFINET for an easy-to-configure, interoperable, Industrial Ethernet network. Oh, we can connect PROFIBUS, too, obviously.
Making a bit of a recovery in favor of fieldbuses, the author notes that interest in Industrial Ethernet and wireless has created a demand for fieldbuses. We’ve observed that also. He quotes ODVA’s Katherine Voss as seeing more hybrid networks. Apparently that means DeviceNet plus Ethernet/IP. The hybrid networks we see include PROFINET and PROFIBUS plus DeviceNet, Modbus, serial, and other protocols.
We can skip over the EDDL and OPC sections; nothing controversial there. But the author’s observation on Ethernet warrants a correction: “Ethernet in industrial networking (IEEE 802.3) has been characterized mostly by the quasi-proprietary icing that many suppliers have spread on top of the basic TCP/IP protocol.”
“Quasi-proprietary” should say “open.” All the Industrial Ethernet organizations have already or are in the process of publishing their open standards through IEC. And the “icing” that he refers to is called an application; you know like SMTP and POP for email, HTTP for web browsing, and CIP and PROFINET for industrial automation. If you don’t have an application, nothing will happen. This is not “icing;” this is what makes the system work!
Wireless is just one monolithic thing he seems to assert. I’ve already ranted about that concept enough.
And now my first flight segment is coming to an end, so I’ll stop here and post this when I hit the ground. Look for more news reviews over the coming week.