ecent developments in the digital fieldbus space have us wondering if the Great Fieldbus Wars of the 1990s are going to flame up again. Emersons Martin Zielinski has been touring media hotspots with the latest on the EDDL standards. The HART Communications Foundation is piggybacking a new user-focused associate membership on the revitalization of the use of HART enabled instruments for asset management and calibration documentation requirements. ModbusTCP has received IEC acceptance as a standard, and the Siemens folks are the fount of a new public relations campaign to introduce Profinet as the one and only fieldbus you will ever need.
EDDL vs. FDT-DTM
The recent decision of Siemens to adopt the EDDL standard instead of the rival FDT-DTM standard for electronic device descriptions means that the majority of the large automation manufacturers have decided that EDDL is superior to FDT. Zielinski made the point that the most significant superiority issue is that FDT-DTM is operating system and software program dependent, while EDDL requires much less overhead to produce the same quality of API. (A copy of Zielinskis white paper on EDDL, "EDDL: The Technology Behind Interoperability," is posted on ControlGlobal.com.) An FDT spokesperson, quoted in Andrew Bonds Industrial Automation Insider, said that counting out FDT/DTM is premature, saying, For your information, there will be a few other big names backing FDT in the coming months, so please do not believe all the âspin at the moment.
Profinet is being rolled out, according to Siemens, over the next few years, with machine control being the first of its functionalities to be offered. Process control with Profinet will await final refinement, and, says Siemens is not going to be available until early in 2006. The Siemens tactical marketing team CONTROL met with in November insists that Profinet for process control is not vaporware; its just not ready yet. Eventually, Profinet will absorb all other Profibus offshoots.
More Fieldbus Wars?
Siemens began a public relations campaign last month to introduce Profinet to North America. Their campaign sounded a little like The Fellowship of the Ring as they talked about one network to rule them all. Is this one of the first shots in a new Fieldbus War?
Not as such, says industry analyst Andrew Bond, because theres not a direct standards issue involved this time. In practice all Industrial Ethernet implementations can, in theory, coexist within the same implementation. So what we are going to see is probably what we should have had first time arounda straight fight to establish market supremacy.
But what does that mean to the typical automation enduser? John Berra, CEO of Emerson Process Management, but wearing his Fieldbus Foundation Chairman hat says, I dont think theres another war, but there will always be controversy. Unfortunately, many customers cant sort this out, and so they stay away from really good technology.
I think the problem is and always has been confusing fieldbus for process with fieldbus for factory automation, continues Andrew Bond, and it's further confused when you add in Ethernet.
Bond points out that Foundation Fieldbus is irrelevant in the Factory Automation market where the issue is Profibus vs. DeviceNet and ControlNet and now Profinet vs. Ethernet IP. That one, he says, divides on Europe vs. North America lines (with the UK as ever trying to make its mind up in between).
Berra agrees. Profibus is a world leader in discrete busses, he says.
On the other hand in the process market, Bond points out,especially oil & gas, refining and petrochem, Profibus is a non starter if you want to do control in the field because it doesn't support it.
According to Bond, FF is the only game in town and that effectively shuts out Profinet as well. As for hybrids, as we've seen, people are quite happy to have Profibus DP for the non process part of the plant - controlling drives etc and FF for the process part.
One fieldbus to rule them all is as much nonsense as Foundation fieldbus for factory automation, Bond concludes.
CONTROL contributor, and Chairman of the Foundation Fieldbus User Group, Ian Verhappen, comments that lack of knowledge is the issue since the Reader's Choice Awards selected Ethernet as communications bus of choice when it only describes the Physical and Data Layers. This round of Fieldbus wars is being fought at the higher level - integrating data with the Enterprise and hence all the Profinet, HSE, and Ethernet/IP discussion. The EDDL/FDT skirmish is part of this battleground.
Verhappen continues, The Field Communications Device to Host battles are completed and each 'bus' has their niches and the cooperative project(s) as exemplified by the EDDL work confirm this. The Network level is where the struggle is now happening and many of the Industrial Ethernets available are all vaporware since few if any host systems support them.
The real problem, Verhappen insists, is that if hosts do support the Industrial Ethernet, it certainly is not in an open and interoperable fashion. End Users would like to see an interoperable control network layer based on standard protocols, he says, but at least for now each manufacturer is protecting that proprietary part of the system.
Verhappen points out the irony in the fact that todays process automation systems are standard on the bottom with FF, Profibus, and HART and standard at the top with web based HMIs on a Windows platform but proprietary in the middle. If they don't keep part of it proprietary they will not be able to sell as many services, Verhappen says, though he comments that this theory should be tested since most companies now rely on their automation supplier or system integrator to do all the system design, installation, commissioning for them anyway because they do not have the skills in house.
Verhappen concludes, If the automation companies are not careful, folks will simply use the Fieldbus of choice at the lower level and then 'standard' Ethernet web servers as the interface to this information and HTML/XML from their on up across multiple enterprises.
The other big player in this set of skirmishes, surprisingly, is the HART Communications Foundation, whose relatively ancient set of protocols was expected to wither away by now. Advances in the standard itself, coupled with the absolutely enormous installed field instrument base have made HART a viable candidate for plant digital communications in the process automation area.
Berra confirms this, now with his Emerson hat on. I have always viewed Foundation Fieldbus and HART as complementary, not competitive, he says. HART has a lot of value, particularly for existing installations. Most of the projects we do have a combination of HART and Foundation Fieldbus.
But Berra insists that HART is, at best, a stopgap. HART has some limits with data rates, and as diagnostics get more sophisticated, the limitations become more apparent.
With Profinet for process automation not yet available (is it vaporware, as Ian Verhappen suggests, or just not ready yet?), Andrew Bond may yet have the last word, at least for now.
On the specifically process front, the key decision to my mind remains whether and when Siemens decides that it has to back Foundation fieldbus in order to become a credible player in the international oil & gas, refining and petrochem markets.