A recent article from McKinsey & Company suggests that AI can be used to accelerate process optimization, though many plants are not ready. Since automation intelligence and informed decision making cannot happen without adequate inputs, McKinsey suggests the first place for improvement is in sensor deployment—in calibrating and cataloging both new and existing instrumentation devices. The McKinsey report found that 75% of plants have instrumentation in place that controls critical process variables, but only 70% of these devices are properly calibrated and cataloged. ^[1]

Self-Calibration Removes Drudge Work for Technicians

Calibrating sensors has come a long way since the days taking them off line and moving them to the lab for checking ice-points and boiling points. Today, temperature sensors (the most measured process variable) can maintain their own internal, accurate and traceable temperature reference points.

The Endress+Hauser iTHERM TrustSens TM372 self-calibrating sanitary RTD sensor helps reduce risk and improve process safety in hygienic applications. Photo courtesy of Endress+Hauser

“Endress+Hauser produces the TrustSens line of self-calibrating temperature transmitters, which use the Curie method with an internal traceable reference point,” says Ola Wesstrom, senior industry manager, food & beverage. “Every time the real temperature crosses the 244°F calibration point, the sensor checks its measured value against the reference. This is far superior to the old method of using dual sensor elements to detect drift because plant personnel could not be sure which sensor was drifting—the primary or the reference. With a Curie ceramic, the reference is permanently stable and traceable.”

While on the subject of temperature, Omega recently announced some new functionality for its HANI clamp sensor, which FE covered in its engineering R&D column. Recently, the temperature transmitter—with analog and smart digital outputs—has been sealed to meet IP67 immersion ratings (1 meter of water for 30 minutes), which means the ability to withstand washdowns, and its temperature range has been extended to -20°C to 100°C. The clamp-on sensor works on industrial and sanitary metal pipes, plastic pipes and metal tanks.

Omega’s HANI clamp-on temperature transmitter can be used conveniently to monitor temperature points in process piping and tanks, and is now able to withstand washdown environments. Photo courtesy of Omega.

Yokogawa’s Sushi Sensor product line plays a critical role in the maintenance of plants and factories, where the data they provide can be combined with automation applications to continuously monitor pumps, motors and other assets, including real-time data on the health of these systems, says Anu Mahesh, industry and product marketing manager at Yokogawa. These capabilities allow end-users to prevent costly downtime, safety issues and loss of production. The real-time data empowers users to move from time-based to condition-based maintenance because they are now able to predict deterioration of assets ahead of time using AI algorithms embedded with automation applications.

“We have three types of IIoT Sushi Sensors for vibration, pressure and temperature measurement,” adds Chris Costlow, Yokogawa industry and product marketing manager. “We also have a wide range of other wired sensors and transmitters that can be used in the food and beverage industry. Additionally, our new flowmeters offer our Total Insight technology, providing inline product health check verifications and diagnostics without powering down or removing the device from the process. These functions help predict future failures by empowering users to monitor remotely and plan service downtimes.”

Yokogawa’s line of Sushi Sensors offers field-proven technologies capable of handling the toughest conditions or application challenges, helping users maximize efficiency, decrease downtime, and pinpoint possible equipment issues more quickly and accurately. Photo courtesy of Yokogawa North America.

Yokogawa also provides vibration sensors, which can be used for maintenance of pumps, motors and other assets—both within and external to the process. These sensors are also self-calibrating, so they do not require any periodic maintenance for calibration purposes.

When Sensors Communicate with Staff

From purely a calibration or ranging point of view, Costlow says it’s not necessary for Sushi sensors to report any data to technicians regarding calibration status. However, AI-based asset management software applications do work in conjunction with the sensors to detect other potential problems ahead of time, alerting the maintenance staff. 

For flowmeter and pressure transmitters that are not self-calibrating, Costlow says users can set alarm thresholds and link to these alarms, which notify them when the instruments are no longer within the process specifications. These alarms inform the user that the instrument requires maintenance or calibration. 

For best results, sensors should indicate issues locally and transmit this information for consumption by a host controller as well as for remote diagnostics, Wesstrom says. For example, a status light on the TrustSens instrument indicates these conditions, and the device also communicates this diagnostic data via HART protocols to a host controller, in addition to calibration information. This creates a traceable calibration record, which can be used for documenting and relaying current information to plant personnel. A user can program the permitted accuracy tolerance to meet a wide range of application-specific requirements.

Keeping track of smart sensors is an easier task than dealing older analog instrumentation. Sensor manufacturers like Yokogawa will offer some method of tracking deployed sensors. For example, a collaborative information (CI) server and asset health management solutions tag and catalog each sensor in the plant and monitor it remotely. All the process and health data from each of these sensors is visible to the operators managing the plant. The supplier’s solutions also provide the ability to set alarms and perform predictive analysis using AI while empowering users to determine when to schedule maintenance/replacement of batteries for wireless sensors.

Use One Sensor/Transmitter to Measure Multiple Variables

Getting the physical sensor count down in a facility is easier today. There are many multivariable sensor options available, Wesstrom says. Coriolis flowmeters have been around for a while with mass flow, volumetric flow, density, concentration—such as °Brix—and even viscosity available. Today, many magnetic flowmeters are adding temperature and conductivity, which can be used for CIP concentration or phase separation detection. Additional recent innovations include free space radar devices with foam detection and alerting. Depending on the number of variables and depth of diagnostics capabilities available, Ethernet/IP, PROFINET and IO-link are the most common protocols for instrumentation communication.

“We do offer sensors and transmitters that provide multiple measurements,” says Yokogawa’s Mahesh. “For example, our Sushi vibration sensor measures both surface temperature, along with vibration in the x, y and z axis. Our temperature sensor provides two distinct temperature measurements from two separate thermocouples.” 

Some pressure transmitters and flowmeters offer multivariable measurements, for example, the EJX910 Pressure Transmitter, Rotamass TI Coriolis flowmeter, and VY Series Vortex flowmeter, Mahesh adds. These devices reduce the instrumentation required, which decreases the total cost of process measurements. Several protocols are available to take advantage of this technology, including Modbus, Profibus and EtherNetIP, to name a few.

The old standby 4-20 mA sensors only provide a single measurement, in all cases, the process variable, Mahesh adds. “Digital sensors and transmitters provide not just process values, but also health of the sensor, to be transmitted simultaneously. This allows users to deploy fewer sensors or transmitters while providing more data and reducing costs.”

Wireless Sensor Networks Extend Range

According to the ARC Advisory Group, two key technologies will extend sensors’ usefulness into the future of aggregating data for optimized processing: wireless sensors and cloud technologies. For temperature sensors, wireless communications can “shorten” the wiring, making readings far more accurate as long sensor wiring resistance increases error. While wireless transmission may not be geared to rapid, continuous updates, it can provide reporting on exception of temperature changes and sensor issues. From ARC’s perspective, the cloud is a critically important technology to manufacturers, since it provides a common, economical, globally accessible repository for plant and equipment data and other infrastructure information. ^[2]

Wireless transmitters, both with ISA100 and WirelessHART communications, have been in use for a few decades, Costlow says. The most recent technology in the industrial space is the use of LoRaWAN, a 915M Hz frequency-based solution that allows small data packets to be transmitted over much longer distances wirelessly. This technology fits in the IIoT category and is the transmission protocol used in new sensor products. ^[3]

There is some movement away from traditional protocols that evolved out of wired technologies, Wesstrom says. In many cases, however, wireless protocols are being selected based on application criteria. For example, wireless technologies like LoRaWAN provide extremely long-range connectivity compared to traditional industrial wireless technologies, while protocols like Bluetooth offer high network speed and reliability. There are certainly instances where secured LTE and even on-premises 5G are being used to support both distance and speed. Independent of the wireless technology type or transport layer, users ultimately need seamless dataflow into their various systems to serve a variety of purposes and stakeholders.

For more on wireless sensor systems, see “Consider Cellular Wireless Networks for Today’s IIoT Systems,” Food Engineering, Sept. 28, 2022. 

Digital Sensor Network Protocols are Changing to More Universal Technologies

For PLC connectivity, multiple protocols are available, including Modbus, Profibus and EtherNetIP, Mahesh says. But Ethernet and IIoT networking are changing the scope of wireless and wired protocols.

“While sensors and transmitters in the past were mostly based on MODBUS protocol, the new IIoT technology has coalesced around MQTT/JSON (JavaScript Object Notation), which makes it very easy to deploy these sensors in a cloud-based architecture,” Mahesh adds.

“In many industrial environments today, wireless devices are often installed alongside traditional wired systems,” Wesstrom says. “It is common to see measurement data and diagnostics from these wireless devices transformed into JSON objects for delivery to a host system via REST (representational state transfer, aka RESTful as an adjective) APIs, alongside information from wired assets via PLC I/O. In both delivery mechanisms, field asset information and process values can be easily moved through data pipelines, for instance, using MQTT objects or OPC Tags.”

To the future and beyond

Considering that an Apple iPhone contains an accelerometer, gyroscope, proximity sensor (typically IR), ambient light sensor, barometer, magnetometer/compass, GPS, touch ID and face ID sensors, camera (optical) and probably others known only to Apple—and that new sensor technologies are making it possible to determine all sorts of quality measurements in food once relegated to our taste, sight and olfactory senses—the sky will be the only limit to the data we collect on food and beverage processing—that sky being the “cloud.”

Resources

^[1] “Using AI to accelerate process optimization: Is your plant ready?” McKinsey & Company, August 2024.

^[2] “Temperature Transmitters,” ARC Advisory Group.

^[3] LoRa Alliance: LoRaWAN, an open standard, enables IoT devices to communicate seamlessly, free of any proprietary technology.