FDT is standardized by IEC 62453, ISA 103, and China National Standard GB/T 29618. The full FDT specifications are publicly available on the FDT Group website.
FDT’s core concept involves two types of software components. One is the DTM (Device Type Manager), which acts like a device driver by encapsulating the features and behaviors of a particular field device (for example a radar level transmitter or a robotic welding tool). The second component, the Frame Application, provides a unified runtime environment for all DTMs. The FDT specification fully defines the interface between the Frame Application and DTM (see figure).
With FDT the properties and behaviors of individual fieldbus protocols are managed separately. Neither the DTM nor the Frame specifies a fieldbus. Within FDT, the fieldbus is treated as an abstraction. This means that the DTM and Frame Application are independent of any fieldbus protocol. It also means that FDT is extensible and can be extended to support any present or future fieldbus or communication protocol.
A technology update, popularly referred to as ”FDT2” has been completed and ARC expects most commercial FDT2 products to appear in 2014. ARC interviewed several suppliers that have released or are in the process of developing FDT2 products. While the company representatives were unwilling to speak “on the record” about products still in development, they were willing to share some of what they had learned from their programs thus far. Those interviewed described many important changes made in FDT2. Here are a few that ARC believes will be most important.
Common Components
FDT2 contains a set of “common components.” These software components serve as reference implementations that developers can use as a complete and fully tested starting point. This will speed up development schedules substantially. The DTM common component implements all the defined FDT functionality. Device suppliers build their own unique device functions within the DTM common component.
A reference Frame Application, which implements all the Frame functionality, is also available. This has been fully tested for interoperability with the reference DTM. The Frame has also been enhanced so that it can easily be embedded in other programming tools, something that at least one major automation supplier with whom ARC spoke is doing for the next release of its control programming environment.
The FDT fieldbus abstraction has also been improved so that it is possible to effectively decouple not just fieldbus technology but also controller programs from fieldbuses. In other words, the controller application does not know (or have to know) which fieldbus protocol is providing data. Rather the controller application receives process data independent of the particular communication protocol.
ARC believes that this level of abstraction will be a boon for FDT, for both process and factory automation applications. It will greatly enhance the ability of control engineers to deliver variants of their designs that use different fieldbus or industrial Ethernet protocols, since they will be able to create a single control program that can use multiple fieldbus technologies without modification.
Backward compatibility
Technology updates are great, but not without an easy migration path. The FDT2 Frame has been designed to support both the new FDT2 DTMs as well as DTMs from the existing FDT1.2. This is a critical point for suppliers and end users. One supplier ARC interviewed has developed several hundred DTMs. This supplier has discontinued all its device-specific proprietary software tools and now uses FDT exclusively to configure and manage its products. The ability to use both old and new DTMs means that end users and suppliers can migrate smoothly to the new technology. The FDT2 Frame can also securely manage the installation of DTMs. DTMs for FDT2 can be securely signed, a big plus for simplifying device management as well enhancing security.
Improved performance
The new FDT2 implementation also shows greatly improved performance. The FDT Group commissioned a series of independent benchmark tests to compare FDT2 performance to FDT1.2 in a number of categories. The software has been improved to reduce memory and CPU use for the most common or resource-consuming FDT tasks. In general, the performance improvements result in an order-of-magnitude reduction in the time required for many tasks (see figure). The improvements derive from improved software design, but also from FDT2 moving to the more modern Microsoft .NET technology from the older COM and ActiveX.
Planning One Deliverable for FDT2 and FDI
In today’s process industry, two distinct technologies (FDT and EDDL) are used for device management. Going forward, FDT and EDDL will be become interoperable through the five-group initiative known as FDI Cooperation, which includes FDT Group, Fieldbus Foundation, HART Communications Foundation, Profibus & Profinet International, and the OPC Foundation. These groups are cooperating to develop technology for interoperability between FDT2 and EDDL. FDI hosts and FDT2 Frame Applications alike will be able to use the resulting FDI Device Packages, which play a similar role to the DTM. The ability for both FDT frame and FDI hosts to use the same Device Packages will allow device suppliers to create a single deliverable for supporting their devices, while providing customers with the choice of using either an FDI or FDT host environment.
ARC spoke with a major FDT device supplier that sells mostly to process industries where HART, FOUNDATION fieldbus, and Profibus are the dominant device networks. Its goal is to develop FDI Device Packages that can be used with both FDT and FDI hosts. This gives its customers the technology choice, and enables it to provide a single package for device support. The supplier reports two challenges with this strategy. First is schedule, since the FDI specification is only due out in 2014. The second challenge is the need to support a broader set of fieldbus technologies than FDI is targeting. For this supplier, fieldbus variety is not a major concern, but for other suppliers now using FDT, support for many field communication protocols is of major importance.
Conclusion
In ARC’s view, the two most compelling features of FDT are its extensibility and its powerful level of abstraction. Extensibility enables FDT to be used with any fieldbus, making FDT equally applicable to process and factory automation. FDT currently supports a dozen different fieldbus and communication protocols including the Common Industrial Protocol (CIP).
FDT’s fieldbus and programming tool abstractions should help simplify solutions for machine and process skid builders. ARC has heard machine builders praise this feature in FDT 1.2 (see this ARC report), but cautioned that it was difficult to achieve. If FDT2 greatly simplifies this ability, these applications will find FDT a most compelling technology across industry.
Author: Harry Forbes, ARC Advisory Group
For further information or to provide feedback on this article, please contact your ARC client manager or the author at [email protected]. ARC Views are published and copyrighted by ARC Advisory Group. The information is proprietary to ARC and no part of it may be reproduced without prior permission from ARC.
