Since 1989, the HART Protocol has focused on simplicity, reliability and utility. One of the keys to the success of HART technology has been the insistence on interoperability and interchangeability. Another has been the insistence of the HART Communication Foundation on backward- compatibility. Each revision of the HART Protocol (the current version is HART 7) has been designed from the beginning to be backward-compatible with every other HART version before it.
While many of the features added in HART 7 are to enable WirelessHARTTM, some of those features, such as "time or condition based alerts," or "report by exception," or the PV trending functions, synchronized sampling and time-stamped data are new and important features that designers and end users can take advantage of, in either a wired HART 7 or WirelessHARTTM network, or both.
It is a significant and unique feature of HART 7 that the same commands, recommended practices, and programming and configuration procedures work for both wired and wireless networks. The knowledge base of end users and designers for working with HART is large and varied, and easily translates to working with WirelessHARTTM as well. No other wireless field device protocol has this integration with wired field sensor networks.
Adding HART 7 and WirelessHARTTM to an Existing Control System
Because of the interoperability, interchangeability and backward-compatibility features of the HART Protocol, engineers and operators may simply add HART 7-enabled wired transmitters to their sensor networks as they have always done, either using the traditional current loop configuration, or the multi-drop digital configuration.
One of the significant changes with HART 7 is the rush by nearly all of the major control system vendors to provide native HART 7 access to their DCS and PLC control systems. This means that all of the features shown in Figure 1 for HART 7 are going to be enabled in nearly every control system. This is a significant enabler alone. Matching this with the other half of the HART 7 release, WirelessHARTTM gives the designer and end user unparalleled ability to monitor process variables and other important information.
Making a Mesh of Things
While the technology of installing and using wired HART transmitters and control valves is reasonably well-known, and well-understood, the nuts and bolts of using WirelessHARTTM devices is considerably less well-understood.
The first part of planning a wireless project is to determine the project scope, just as it is when planning a wired sensor network. There are differences, though.
First, you need to determine what a Process Unit is so that you can organize the Wireless Field Devices (WFD) and their network. This maintains existing workflows, data flows and operator responsibilities, and provides for scalability in the future. Determining a Process Unit also sets expectations for the range of the wireless network (at this point, typically less than approximately 250 m).
Most WFD are going to be mounted below the normal obstruction height. That's generally the airspace above the process vessels and tankage, where LOS (line of sight) is located. LOS is defined as "obstruction height + 2 m." Mounting network gateways and repeaters at LOS height will produce maximum range.
Forget the Surveys
You don't need a traditional radio survey, whether you are on a brownfield or greenfield site. Get yourself a scale drawing or an aerial photograph. Printing out images from Google Earth or other aerial mapping programs can be quite effective. You can use scale drawings for P&ID, ILP and asset management for this purpose too.
From experience, we already know that a minimum of five WFD should be used to create a network. The more WFD you have, the stronger the network becomes, because the WirelessHARTTM network is a self-healing, self-organizing mesh. The more sensors, the more paths the signal can go to reach the gateway, and from the gateway to the control system. On your scale drawing or photograph, place the WFDs as if they were standard wired transmitters and final control elements.
In greenfield applications, you can plan to install them as if they were wired—just leave the wires off the drawings. Each field device should have at least three neighbors. If the congestion of tankage and other obstructions is heavy, locate your transmitters within 30 m of their neighbors. For medium obstructions, use 75 m, and for light obstruction, light environmental density, you can go as far as 150 m.
Mount the antennas at least 0.5 m away from any vertical surface, and at least 1.5 m off the ground.
One of the uses of the WirelessHARTTM adapters being supplied by at least five vendors is to permit inexpensive remote location of the wireless antenna, since the adapter can be located anywhere along the HART-enabled current loop. Thus, you do not have to run expensive antenna cable—just run a standard current loop out of the transmitter and use a WirelessHARTTM adapter.
A Process Unit with at least 25 WFD provides saturation coverage of the unit, with minimal need for design recommendations. Locate the gateway in the center of the initial network, or in the center of the Process Unit, with its antenna at LOS height (obstruction height + 2 m), and with direct wireless connection to at least 25% of the WFD in the network. To fortify a network, add more WFDs. WFDs can be used as repeaters. That's a quick fix for fortifying the network during design, or after installation.
Your existing network, once formed, is the foundation for scalability, whether you add WFDs to the perimeter of the network, or to the interior of the existing network. Don't forget that this is a mesh network. Devices do not need to be within the range of the gateway, they just need to be close to another wireless instrument.