What Would You Measure If It Cost Nothing?

Today we use cameras differently from a few years ago because taking a digital photo cost essentially nothing. In the past taking a photo was expensive: the cost of the film, and cost of developing the print. Because today taking a digital photo is essentially free we have completely changed how we take photographs. In the past it was birthdays and holidays only, today we use our digital smart phone to take several digital photos every day; of the shopping mall map, where we parked the car, and of a product label etc. This helps us tremendously in our daily activities. This is the power of digital devices, and the reason why all devices around us are now turning digital. What if measurements were like that? How many sensors would you add to a plant? 1,000? 10,000? Maybe 20,000? Here are my personal thoughts:

Adding a measurement in most plants today would be very costly so it doesn’t get done even if that measurement would help operations & maintenance in their daily work. Never mind the price of the transmitter. Consider the cost of the cable back to the equipment room, the cost of the analog input card point, the marshalling wiring, and in some installations a safety barrier. The cost of the signal ends up higher than the cost of the transmitter. The transmitter is a smaller part of the total installed cost for the measurement. Even if spare multi-core cable pairs, AI card channels, and safety barrier already exists, using them still has a cost because at some point the hardwired system will run out of spare capacity and more has to be added, and the cost of laying cable and installing field junction boxes in an existing plant is very high. If cable trays also have to be added it becomes even costlier. The limits of hardwiring are indeed very ‘hard’. A 12-pair cable can only handle twelve signals, no more. An 8 channel analog input card can only handle eight 4-20 mA signals, no more. For this reason many good plant improvement ideas have been rejected. Moreover, the transmitter may perform multiple measurements such as density and temperature in addition to mass flow, or interface level in addition to level but only the primary measurement gets wired up due to the high cost of analog signals, thereby only using part of the device functionality, so the plant only benefits from part of the device functionality.

With digital fieldbus and wireless the signals are virtual; that is, one signal does not require one pair of wires, one I/O card, and one barrier. One pair of wires and a single fieldbus interface port can handle up to 16 devices and nearly 50 signals. For WirelessHART there are no wires or barrier at all, just a wireless gateway which can handle 100 devices. Note that other wireless sensor networks require far more complex network infrastructure than WirelessHART. The measurement signal is therefore essentially free. Sure, you have to pay for the transmitter but the transmitter is the smaller part of the total installed cost. A wireless transmitter needs no wiring so it is intuitive that the measurement signal is essentially free. For fieldbus measurements the signals are essentially also free. That may not be as intuitive. Here’s why: for fieldbus the number of junction boxes (JB) is pretty much sized based on the number of control loops because most plants are designed with a max of 4 loops per bus (per JB). 4 loops translate into 8 devices (4 transmitters plus 4 valves – less if fewer loops are loaded). That is, each little field JB will have 8 devices or less dedicated to control which pays for the installation up to the JB, the JB can then be filled with other transmitters for monitoring at little marginal cost. Since a digital bus can handle up to 16 devices, all the signals for the additional 8 transmitters or more on the JB are essentially free. That is, both digital fieldbus and wireless make it economical to add lots more transmitters in the plant to improve operations & maintenance. It also means that both digital fieldbus and wireless have great spare capacity to add more instruments at will. The bus may initially be loaded with 10 devices, but this can be increased to 12, 14, or 16 devices, response time permitting. A full 16 devices could mean almost 50 signals in a small JB on a single pair of wires.

Keep in mind that in the world of digital communication one device is not the same as one signal. It is not a 1:1 relationship. Most intelligent devices have multiple sensors and internally calculated variables which are useful to operations, communicated wirelessly or over fieldbus. Because additional signals in a device are essentially free, all the signals in a device can be integrated to the digital system, for the plant to utilize the full device capability. This was not practical with hardwired signals.

Another observation is that 4-input and 8-input digital wireless and fieldbus temperature transmitters respectively also take the place of multiple transmitters, not just multiple AI cards and cabling.

Two-wire differential pressure (DP) transmitters that also measure static pressure and process temperature have existed for a long time, but digital communication enable many more such devices. In the future it appears inevitable that Coriolis flow meters, vortex flow meters, radar level transmitters, and guided wave radar level transmitters will all also measure pressure and temperature. A two-wire multi-variable tank gauging system already exists, so why not a two-wire condition monitoring system, or custody transfer flow metering system. Digital technology makes it possible, and control systems built on a digital ecosystem are ready to accept such devices today.

With almost free measurements, the plant can start adding sensors used in new ways measuring things that were not measured before. Most of these applications are beyond the P&ID. For example;

• Pump & mechanical seal condition monitoring
• Blower condition monitoring
• Air cooled heat exchanger condition and performance monitoring
• Vibration not only on critical compressors and turbines but also on smaller compressors and pumps
• Cooling tower condition and performance monitoring
• Pipe & vessel corrosion
• Eliminate manual reliability data collection: vibration, temperature, and acoustic
• Steam trap failure monitoring
• Relief valve seat passing
• Energy management: water, air (compressed), gas/fuel, electricity, and steam
• Emergency safety shower and eyewash activation
• Manual and bypass valve position
• Relief valve and rupture disk release
• Shutdown valve position confirmation
• Hydrocarbon leak/spill detection
• Cooling water effluent temperature
• Storage tank overfill/spill, roof tilt, water pooling
• Eliminate manual operation data collection: gauges, sight glasses, VA meters

In some countries plants can add hundreds of sensors without upfront investment, instead paying for periodic reports in a multi-year remote monitoring service contract. For instance, plants in Singapore have acoustic sensors on their steam traps monitored by analytics software on a virtual machine in the cloud. Failure reports are received daily or weekly, and energy savings reports are received monthly. Thanks to a mobile network data backhaul, no connection to the plant’s own control system is required. This is the essence of the Industrial Internet of Things (IIoT).

Another obstacle to additional measurements in the past is the costly mechanical installation requiring cutting, welding, or drilling which in turn requires a shutdown. Therefore a key enabler for essentially free measurements is that many of the new generation sensors used in applications beyond the P&ID are non-intrusive. This includes for instance clamp-on temperature sensors, strap-on acoustic sensor for steam trap and relief valve monitoring, bolt-on position transmitter for valves and dampers, stick-on or screw-on vibration sensors, DP flow meter slip-in between flanges, pressure transmitters take the place of mechanical dial gauges reusing existing process connection, and hydrocarbon leak detection sensor lay on the ground or float on water. Plants also fill their existing thermowells with a sensor and transmitter, such as the empty thermowell often found on reactors and columns or between heat exchanger bundles. See further explanation of these applications in this article:

https://www.ceasiamag.com/2014/09/beyond-the-control-room/10658/

It is the first device that pays for the fieldbus cable or wireless gateway, additional devices have little marginal cost. What would you measure in your plant if it was essentially free like digital photos? Well, that’s my personal opinion. If you are interested in how the digital ecosystem is transforming process automation click “Follow” by my photo. Let me know what you think by providing your feedback below, and click “Like” if you found this useful.