Observing Protocol: What’s in Store for BAS?

By Barbara Horwitz-Bennett, Contributing Editor May 1, 2005

Integration of BACnet and LON protocols has become easier, say the experts, but the focus has shifted to the convergence of BAS with IT networks, XML and web services. (Click here for part 2 of this roundtable discussion.)

CSE: There were a lot of products at the AHR Expo this year that claimed the capability to work with both BACnet and LON. Has integration of the two protocols gotten any easier?

HUSTON: I’d say so. Integration has become significantly easier with the introduction of third-party products that speak to both protocols, while providing enterprise connectivity and browser-based access. These products work well and have greatly advanced the realization of open systems.

HOFFMANN: Also, if a manufacturer doesn’t say it has capabilities to integrate both, this will lead to limited credibility in the marketplace.

WICHENKO: There are now formal mechanisms—such as the BACnet Interoperability Building Blocks (BIBBs)—to test for interoperability. Test procedures and gateways have dramatically reduced our concerns.

TENNEFOSS: There are gateways available from multiple companies that allow LonWorks and BACnet systems to share data. However, the problem with gateways is that they don’t permit the full range of services to be supported. That is, a common LonWorks installation tool that is used to install devices from hundreds of LonWorks suppliers is unable to jump the LonWorks-to-BACnet gateway and also install BACnet devices. A flat system architecture—one in which every device can communicate with every other device, with no gateways and a common installation tool—is still the best system architecture. As things now stand, gateways simply pass status information, but cannot support the full range of available services.

CSE: What other technologies are affecting the future of BAS, and what specific effects might they have on BACnet, LON or other proprietary protocols?

HOFFMANN: Convergence with IT networks is the biggest technology trend. It allows us to embed intelligence at lower levels of the system architecture and provides us with a natural bridge to technologies that will revolutionize the way we do business. I believe wireless communication is one such technology. For example, think of a system where all components are enabled for wireless communication across the building infrastructure and shipped complete with controls to the project site. The mechanical or electrical contractor installs the equipment and wires power to it, and the controls come along for the ride. There is no control wiring and, if done correctly, no commissioning.

WICHENKO: There is a new draft standard called BACnet/WS (web services) that has already been through public review. This emerging standard provides robust support for web services that are vital to integrating BACnet-based building systems into the enterprise-wide management-information system.

The BACnet/WS approach is equally suited to integrating both BACnet and LON systems into the enterprise-wide systems to report performance, operating costs and other information to decision makers.

At the same time, tumbling prices of Internet Protocol (IP) infrastructure has accelerated implementation of IP technology into small, simple devices. BACnet’s robust model for using IP in a native fashion, rather than the weak tunneling method used in CEA/EIA 852 for LON systems, makes BACnet the stronger option to take advantage of this new technology.

EHRLICH: In addition to the Internet-based solutions, the other giant technology impact will be from many of the new wireless mesh solutions that are now becoming available. These solutions allow for very economical installation of devices without wires. The solutions offer long battery life and reliable communications. Since communications wiring for sensors and devices is often 30% to 50% or more of project costs, the use of wireless technology could have a giant impact.

HUSTON: Wireless devices and mesh networking will greatly enhance the BAS application base, while reducing in-stalled costs for BACnet, LonWorks and proprietary systems. Many promising wireless products have hit the market recently offering a wide variety of features and benefits from single-point sensing to broadband transmission. This promising technology has been a long time in coming and has evolved enough to become a reality in the near future.

TENNEFOSS: WiFi and LAN technologies are making it easier to seamlessly blend control and data systems, leveraging the strengths of each to achieve LonWorks automation systems that are easier to deploy and maintain. Looking forward, power-line signaling—the ability to communicate through the AC-power circuits without any new wiring—will play a profound role in reducing the installation and ongoing cost of LonWorks emergency lighting, air-quality, security, power sub-metering and lighting systems. New self-installation technology—the ability to create an ad hoc network without using any tools—will also play a role in simplifying the deployment of BAS. Finally, advancements in semiconductor fabrication will make it possible to network even the lowest-priced sensors and actuators.

CSE: Where does XML fit into this whole equation?

HUSTON: XML and web services will be the next major evolution for integration of building subsystems and the sharing of building data with external users. They will provide advanced integration between building systems to permit a straightforward methodology for exchanging basic information among traditionally stand-alone systems such as door access control, fire alarms and elevators. XML groups are currently developing schemas and interfaces to simplify implementation. As all building subsystems migrate to Internet Protocol (IP) based networks as a means of control and data communications, XML and web services will provide the path to greatly enhanced efficiencies, intelligent building applications and new opportunities for system integrators.

EHRLICH: XML-based standards, such as BACnet/WS and oBIX, will quickly become expectations for building information that is shared over the IP network. But will it replace other solutions such as BACnet and LonTalk? Not in the immediate future, because it is not economical. But we will see systems starting to connect, followed by large devices, and in time, perhaps five to 10 years, all devices will have direct network connectivity.

TENNEFOSS: To Paul’s point, XML is not a communications protocol. Therefore, it is not suitable for non-intensive data applications like control devices, and it does not incorporate network installation or network management capabilities. XML is not a panacea that will free owners from proprietary systems. Closed implementations can serve as Trojan horses under the guise of an open IT standard, luring unsuspecting specifiers and owners into the trap of a sole-sourced system. oBIX can help to ameliorate this issue by corralling the Trojan horses through XML object definition, a process that has started but needs to gain momentum to be effective.

WICHENKO: XML will be the next technology to be adopted by the controls industry, and BACnet’s object model is ideally suited to migration to XML. As a matter of fact, the BACnet XML Working Group has been working on various initiatives that would exploit XML for use as a configuration and object-description vehicle.

CSE: Is interoperability, in general, just a wishful panacea, or is real progress being made? If so, can you provide a couple of examples?

HOFFMANN: No, open protocols are not the miracle medicine for all that ails the building-systems market, but they certainly are a wonder drug. Five years ago, if I wanted to select a variable-frequency drive for a project before the automation system was awarded, I had to take my chances. Now, by specifying BACnet or LonMark devices, I can feel secure and protected in my decision. The new terminal at the Toronto International Airport is a fine example. The owner asked for LonWorks networks at the field level and XML integration between some of the enterprise, airport-specific systems. We were able to supply a wide variety of products and integrations based upon this specification. The building-controls system at Camp Pendleton in San Diego is another poster child for standard protocols. The original workstation and LonWorks networks continue to operate, and as buildings are upgraded and integrated, the contract specifies compatibility. The building system is then integrated seamlessly into the central monitoring and control network.

HUSTON: Real progress is being made daily. Our business unit within Teng concentrates 100% on the application of open standards, interoperability and enterprise connectivity. It’s all about getting information into the hands of those who need it, in a cost-effective way. LonWorks, BACnet, MODBUS, TCP/IP and others are providing the tools we need to obtain the results our clients have grown to expect. Examples include the city of Chicago’s Global Building Management System, the General Services Administration, Boeing and a variety of major telecommunications companies.

EHRLICH: It is important to define what “interoperability” is. It is the ability to connect systems or devices together. And it covers a spectrum—from integration on one end to interchangeability on the other. With integration it is possible to connect two devices, although it may not be a trivial task. With interchangeability a technician can readily swap devices and expect them to work. We are way past the simple ability to integrate. We’re moving in the direction of interchangeability. For example, many projects integrate large pieces of mechanical equipment such as boilers and chillers, and it is more typical than not to see lighting and HVAC integrated.

TENNEFOSS: Interoperability is a reality in the LonWorks community, and there are literally thousands of buildings running today with multi-vendor, flat, open, interoperable control systems. Engineers and owners have the right to expect that when they specify or buy an interoperable control system, they will receive the benefits of interoperability—multi-vendor systems for which the design, installation, maintenance and service components and contracts can all be openly bid by multiple sources. The LonWorks community has an excess of 500 certified products, 62 functional profiles, 225 network management tool plug-ins and more than 40 million device-level shipped projects.

For example, Tokyo’s Rappongi Hills, owned by Mori, consists of 170,000 monitored points spread among 16,500 interoperable LonWorks devices from multiple manufacturers, and uses IP backbones integrated with LonWorks twisted-pair networks. In saves 20% per year in HVAC-related costs, while allowing the owner to open-bid all adds, moves, changes and service contracts.

Owners who expect interoperability and receive partially interoperable systems have a lot to lose in the process. Partial interoperability is akin to no interoperability because it obviates multi-vendor systems for which the design, installation, maintenance and service components and contracts can all be openly bid by multiple sources. A consulting engineer should not have to understand BIBBs and PIC (protocol implementation conformance) statements to know whether a product can communicate with what, for all intents and purposes, looks like another identical device.



for part 2 of this roundtable discussion.

Weighing the Merits of LON and BACnet Is it fair to say that LON is perceived as the superior field-level system, but BACnet provides superior capabilities as far as alarming, scheduling and trending?
“Technically, both BACnet and LonTalk are similar in what they accomplish,” says Paul Ehrlich, president of Building Intelligence Group, White Bear Lake, Minn. “I would not characterize either as being superior. Frankly, integration with either technology remains challenging.”
Terry Hoffmann, manager of global products marketing for Milwaukee-based Johnson Controls concurs—to a point. “There are many projects on both sides of the protocol fence, but in general, there appear to be more and better standards for field devices on the LonMark side,” he says. “More attention is being paid to the building- and enterprise-level side with the BACnet standards. But the needs of the job trump the individual standards.”
John Huston, P.E., senior associate and director of technology integration with Teng & Assocs., Chicago, has a slightly different take. “LonWorks provides superior field-level integration capabilities between multiple manufacturers for a number of reasons, including the consistency of application by manufacturers, availability of third-party network-management tools and third-party options for the development of graphical user interfaces and enterprise connectivity,” says Huston.
On the other hand, he suggests, BACnet integration is reduced to the lowest common denominator, typically the head-end workstation or supervisory controller. He argues that since alarming, scheduling and trending are primarily carried out by the supervisory controller or head-end workstation, BACnet is thought to have superior capabilities in this realm.
“But how valuable is that if it can only provide those features for a single product line?” he asks. “Device-level integration of similar equipment—e.g., VAV box controllers—is required to realize the benenfits of open systems.”


Paul Ehrlich , President, Building Intelligence Group, White Bear Lake, Minn.

Terry W. Hoffmann , Manager, Global Products Marketing Johnson Controls, Milwaukee

John Huston , P.E. , Senior Associate, Director of Technology Integration, Teng & Assocs., Chicago

Michael Tennefoss , Vice President of Marketing Echelon Corporation, San Jose, Calif.

Grant Wichenko , P. Eng. , Appin Assocs., Winnipeg, Manitoba, Canada