By Dave Harrold
To outsiders, the New Sampling/Sensor Initiative (NeSSI) may look like one of those ideas whose time is never going to come—a good idea that, for whatever reason, never quite catches on. Those closer to its development and its early adopters see things differently. The road to wide NeSSI adoption has been a little rougher than may had hoped, but over time, improvements and new developments have kept alive the hope that it will finally fulfill its potential.
First, a bit of history. Near the end of the 20th Century, analytical measurement engineers recognized that, while advances in computer hardware and software allowed analytical devices to become more capable, more reliable and easier to use, the basic design of sample-handling and delivery systems hadn’t kept pace. Convinced that improvements in sample-handling systems would allow placement of analyzers “at process,” and with sponsorship from the Center for Process Analytical Chemistry (CPAC) at the University of Washington in Seattle, a dedicated group of end users, equipment manufacturers and academies launched NeSSI in 1999.
From its inception, NeSSI’s objective has been to create an open architecture platform on which manufacturers and end users could assemble miniature, modular, intelligent sampling systems that: improve analyzer system performance; reduce design, build and installation costs; and reduce operating and maintenance costs.
Achieving these goals has proven to be a bit more complicated than first thought. Rick Ales, NeSSI’s secretary and a marketing manager at Swagelok, says, “NeSSI can be thought of as a ‘two-rail concept.’ One rail provides the fluid interface, and the other rail provides the electrical interface. Sample systems that implement the fluid interface (mechanical rail) as defined in ANSI/ISA 76.00.02 are considered NeSSI Generation I systems. Adding the NeSSI bus interface (electrical rail) to a Generation I system creates a Generation II system.”
While progress and success were achieved on the fluid interface (mechanical) portion, NeSSI’s steering committee was surprised when it turned its attention to the bus interface (electrical) portion of its initiative. “When we released Generation II’s specification in 2002,” says Dave Veltkamp, senior research scientist at CPAC, “we thought the steering committee could simply tell people what communications network to use, and that all NeSSI developers would use it. However, once we looked at what was available, we found that nothing met all of NeSSI’s requirements.”
That revelation required the steering committee to take a more pragmatic approach to establishing NeSSI’s bus technology—one that allowed the users and vendors to decide cooperatively what bus communication protocol was best, with the component vendors agreeing to support whatever protocol was eventually chosen.
From the beginning, the Controller Area Network (CAN) bus protocol had been the preferred technology, but CAN bus did not meet the NeSSI Generation II requirement for intrinsic safety. Then, in 2004, two significant milestones were achieved. First, the technical committee for sensor technology of the Instrumentation and Measurement Society sponsored draft standard IEEE P1451.6, producing an intrinsically safe CAN bus solution, which met NeSSI’s intrinsically safe requirements. Second, the NeSSI Generation II spec was modified to include Foundation fieldbus as an acceptable NeSSI bus technology—a definite benefit for new installations, but not so much for existing facilities with cable trays full of 4-20mA wire.
Wiser from this experience, NeSSI’s steering committee turned its attention to how best to achieve interoperability, maintain its open architecture goal, and accommodate existing hardwired facilities.
A key technology used by both 4-20 mA and fieldbus-based instrumentation is Electronic Device Description Language (EDDL). EDDL has been adopted by ANSI/ISA, and is part of the international consensus standards IEC 61804-3 and -4. Therefore, EDDL has a high acceptance rating in the instrumentation and controls industry.
Terry Blevins, of Emerson Process Management and the ISA SP104 committee’s chairperson, says, “The main reason for NeSSI’s interest in EDDL is that it enables manufacturers to describe the features of their sampling systems. Because EDDL is supported by all major control system manufacturers, sampling system manufacturers only need to develop one EDD file, which makes it easy for them and end users. Also, because the EDD file is a text file and is independent of any operating system, the original EDD file can coexist in a control system, even if the sample system manufacturer later updates its product and EDD file with new features. Finally, EDDL enhancements that were added a few years ago and that are reflected in the latest IEC 61804 standards are fully capable of supporting highly complex sample system and ‘at process’ analyzer applications with effective graphic interfaces.”
Another piece of the interoperability puzzle is HART, which is the most recognized and successful deployment of EDDL, with more than 20 million installed HART-based devices worldwide. Developed in the late 1980s, HART is based on the Bell 202 Frequency Shift Keying (FSK) standard that defines how to superimpose low-level digital communication signals on the same wire as 4-20mA analog signals, which satisfies the needs of most hardwired plants to use at-process instrumentation.
In 2007, the HART Communication Foundation released its WirelessHART standard and stated that wired and wireless HART devices can coexist on the same network. Though at press time NeSSI’s steering committee had made no announcement about including EDDL and HART in its Gen II specification, it seems reasonable to think that both of these technologies will play a major role in helping NeSSI achieve its goal of producing cost-effective, open architecture sampling systems.