Independent protection layer issues offer second front

Excerpts from the April 2007 issue of Industrial Automation Insider newsletter, Andrew Bond, editor. For more information, go to www.iainsider.co.uk. Reused with permission.

By Andrew Bond

Share Print Related RSS
Page 2 of 4 1 | 2 | 3 | 4 View on one page

However, McCrae-Steele argues, if the BPCS and the SIS are based on the same hardware platform, they could suffer a common cause or systematic failure. Consequently, an analysis which results in SIL 1 and SIL 2 safety-instrumented functions (SIFs) based on the assumption that the BPCS and DCS are providing IPLs, should end up with mostly SIL 2 and SIL 3 SIFs when the common platform is taken into account. Moreover, one with SIL 3 requirements would be raised to SIL 4, which would send the whole scheme back to the drawing board.

Cyber Security
McCrae–Steele’s other major concern centers on cyber security and the possibility that closer integration with the DCS and hence, potentially, with the outside world, opens the safety system up to the risk of malicious attack. A process plant manager’s worst nightmare, he suggests, is that of “A DCS-embedded safety system where a malicious cyber attacker penetrates the firewalls of the control system connected to the site LAN or corporate WAN, disables the SIS, and uses the DCS to blow up the plant.” However, having argued that “The safety system, as the last line of defense, needs to be protected,” he seems to undermine his own position by suggesting that the solution is “smart integration at the information, configuration, asset management and HMI levels,” which sounds a lot like the current Triconex solution. Given that, as he suggests, “all systems are vulnerable,” it is difficult to see how “smart integration” renders the safety system any less vulnerable to malicious cyber attack than any other form of integration.

It is clear that Triconex is opening up a new front in its battle with the integrated SIS lobby and one which may give some potential users of integrated systems pause for thought. Whether, by adopting a position which some may deem over-alarmist, McCrae-Steele has weakened rather than strengthened the case for independent systems will be for others to judge. It is unfortunate that, by canceling its Galveston symposium, TÜV has removed the opportunity for an early debate on the issues raised.

Are FDT and EDDL on Course for Convergence?
With representation on the FDT Joint Interest Group, the Fieldbus Foundation and Profibus International, ABB is probably in as good a position as anyone to assess the current state of play between the advocates of FDT/DTM and EDDL. Officially the ABB position is that expressed by Instrumentation Group vice president of Technology Sean Keeping, namely that EDDL and FDT/DTM are complementary rather than competitive. “We support both, and both have their uses in our equipment and systems,” he told the ‘What’s New in Instrumentation’ workshop at ABB Automation World. “We don’t see one dying and the other winning.”

However when we asked him whether there were now signs of movement towards some form of convergence or even a merging of FDT and EDDL, he said that this was indeed being discussed but there was as yet no timescale for such a move. Keeping, who is close to the FDT Group, was confirming what we had already heard from ABB’s senior vice president, Global Marketing Group, Mark Taft, namely that there are signs of a rapprochement between the FDT and EDDL camps. Taft sits on the Fieldbus Foundation board of directors and might perhaps therefore be expected to be more aware of EDDL thinking.

Joint Panel
More details of these developments may emerge later this month when the EDDL Cooperation Team and the FDT Group are jointly hosting an “Editor Panel” at Hannover Fair. The EDDL Cooperation Team brings together the HART Communications Foundation, the Fieldbus Foundation, the OPC Foundation and the Profibus Nutzer Organization and was responsible for the development of a single EDDL specification which is common to HART, Profibus and Foundation fieldbus. Interestingly, the invitation came from John Weet of PR agency HHC Lewis whose principal client is Emerson Process Management.

We asked Weet if he could comment on whether the forthcoming event had anything to do with rumors of the hatchet being buried between EDDL and FDT, but he played the suggestion back with a straight bat, saying that he couldn’t make any comment ahead of the event, but that he would “keep you in touch with what is going on as soon as I am authorized to do so.”

These developments, if developments they be, come at a time when rumors have also been circulating in Europe of problems with interoperability testing of FDT/DTM-based devices.

The latest figures from Profibus International show that a total of 3.4 million Profibus nodes were sold in 2006. This, it is claimed – although it’s surely not necessarily the same thing – brings the total number of Profibus nodes installed world-wide to 18.8 million, making it almost certain that the target of 20 million nodes by 2008 set in 2004 will be comfortably exceeded. “We actually sold more nodes last year than any other fieldbus organization …,” observed North American PTO executive director Mike Bryant, before claiming rather more contentiously that “The pattern set in discrete automation was repeated in process automation, and Profibus remains by far the strongest candidate for fieldbus applications in both market sectors.”

As Control magazine’s Walt Boyes helpfully points out in his "SoundOff" blog, both Profibus and Foundation fieldbus can only make their respective claims about installed nodes by pretending that HART isn’t a fieldbus.

Fieldbus or no, it’s giving most users what they’re looking for, which is remote diagnostics and integration with their asset management solutions. Current estimates put the HART device population at 22 million and, with Wireless HART now close to becoming a reality, a further surge seems a racing certainty. Set against that, PI’s proud boast of 630,000 installed PA devices by the end of 2006 looks pretty meager and isn’t made much more impressive by adding in all the non-PA devices in process applications which, it says, brings the total to 3.3 million. That, it claims, is five times more than any other fieldbus but, as no doubt Boyes would point out, it’s still just 15% of the installed HART population.

Page 2 of 4 1 | 2 | 3 | 4 View on one page
Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments