JAN-FEB 2019

Issue link:

Contents of this Issue


Page 27 of 55

28 INTECH JANUARY/FEBRUARY 2019 WWW.ISA.ORG SYSTEM INTEGRATION some cases contain web servers and Ethernet ports for directly connecting to the Internet. Smart instruments can acquire so much data, they need a high-speed digi- tal interface to send it all. For example, some Coriolis flowmeters can simultane- ously detect multiple measured process values, including mass flow, volume flow, density, concentration, and temperature. In addition to these measured variables, built-in electronics monitor instrument performance and report status and di ag- nostic values. Once smart instruments became widely accepted, mostly by add- ing the aforementioned features to the transmitter, smart sensors followed. Smart sensors improve operations Because digital information resides in the sensor and is communicated to the transmitter, health diagnostics can be per - formed, and the state of the sensor and transmitter health can be communicated to the host systems in real time. Real-time diagnostics and sensor-health data allow personnel to better manage a sensor. The need to clean and calibrate the device can be proactively managed, rather than reac - tively performed. In fact, some smart sen- sors can determine if they actually need to be cleaned and calibrated. For example, calibration cycles for standard temperature sensors in criti- cal service are every six to 12 months. This requires a technician to remove the sensor, take it to a lab for calibra- tion, and then reinstall it. But one re sistance temperature detector (RTD) sensor is able to determine if it needs a calibration when used in sterilize-in- place (SIP) processes (figure 2). In SIP processes, steam at 121°C (250°F) is used to sterilize equipment. The sensor uses a reference material with a Curie point of 118°C (244°F). When the SIP pro- cess reaches 118°C, the reference sensor sends a signal. Simultaneously, the RTD measures the temperature. Comparison between these two values is used to de- termine if the temperature sensor needs calibration. If both sensors read a value close enough to 118°C, the RTD sensor is still in calibration. Another example of real-time diag- nostics and sensor-health data is a four- pole conductivity sensor (figure 3). This sensor uses four conductors to measure con- ductivity, and its four-pole design allows the sensor to operate over a broader range of measurement than two-pole con- ductive sensors. It uses digital sensor technology in the head of the sensor to digitize the measurement signal and pro- vide a host of performance and diagnostic information. One diagnostic function that makes this sensor particularly smart is electrode connection surveillance, which monitors the connection between the electrodes and the electronics. If there is a connec- tion error, an error message is sent to the transmitter to notify the user of a con- nection problem within the sensor. Accessing smart sensors Maintenance personnel are stretched thin at many process plants and facilities, increasing the need for digital technolo- gies, such as remote access to an instru- ment beyond the control system. By using digital communications, especially over an industrial Ethernet network, an instru- ment can become a "thing" in the Indus- trial Internet of Things. Some more sophisticated digital trans- mitters have an embedded web server, permitting properly authorized access from any device connected to the Inter- net and capable of hosting a web browser, such as a smartphone (figure 4, page 24). Two of the leading networks for local access to smart instruments from host systems are Modbus and EtherNet/IP. Common hosts are control systems and asset management systems. Modbus is an open protocol, allowing any manufacturer to integrate the pro- to col into an instrument. Modbus is a serial, master-slave, protocol. The master requests information, and the slaves re spond, with one master communicat- ing with up to 247 slaves. Each slave in the network is assigned a unique ID. When a Modbus master requests information from a slave, the first data communicated is the slave ID. Modbus can be difficult because one can use 16- or 32-bit signed and unsigned integers, ASCII strings; discrete on/off val- ues, and 32-bit floating point numbers. To program a system for a device using Modbus communications, a significant amount of information is required about the slave device and its registers. A pro- grammer has to obtain a Modbus map from an instrument manufacturer and carefully program the master to commu- nicate properly with each slave device. An improved version called Modbus TCP/IP is now available, whereby Modbus data can be framed in a TCP/IP packet, al- lowing the information to be more easily communicated over an Ethernet network. EtherNet/IP is becoming one of the most widely used industrial protocols due to its ease of integration and operation. Like Modbus TCP/IP, EtherNet/IP data is transferred in a TCP/IP packet. Each de- vice on an EtherNet/IP network presents its data to the network as a series of data values called attributes. Figure 3. Technology is enabling real-time diagnostics and sensor-health data embedded in sensors. For example, the CLS82D four-pole conductivity sensor incorporates smart diagnostics. Figure 2. The TrustSens RTD checks its calibration every SIP operation.

Articles in this issue

Links on this page

Archives of this issue

view archives of InTech - JAN-FEB 2019