Selecting appropriate equipment for a new or improved process is challenging, especially when it involves unfamiliar areas. For example, someone with experience in flow or pressure measurement might find temperature sensors daunting. How does one approach a new application area? Equipment selection begins with a careful analysis of the process and the controlled objects. Every measurement point must have a purpose. In larger control plans, vague purposes can easily lead to incorrect equipment selection. Overly precise equipment wastes money, while overly simple equipment fails to provide sufficient accuracy, sometimes even providing incorrect information. Carefully examine each point on the flow chart and ask the following questions: ■ What are the measured variables and their corresponding units (pressure, temperature, etc.)? ■ What is the operating range of this process when it is operating normally? This question is much harder to answer than it sounds if the process is entirely new or still in the design phase. ■ What is the abnormal operating range of the system when it is in a faulty state? To protect the equipment in the system from damage, by what extent should the operating range of these devices be extended? To prevent abnormal peaks in the workflow from damaging equipment, high-level protection measures are needed, which reduces the accuracy level. ■ How high should the accuracy of the measurements be? This has a significant impact on the cost of the equipment. ■ Besides being a method of representation, do units serve other functions? Greg Ferro, a systems integration engineer at AmeriChem Systems, says, “Your processes rarely operate under extreme conditions. They usually have very narrow operating ranges, and you just need to choose equipment that can meet those ranges. Of course, sometimes engineers can't predict actual operating conditions, and this unpredictability is higher than people think. If a customer asks to design a system that can process 80 gallons per minute, but you find that the system actually only needs to process 8 gallons per minute, then you're in trouble.” Choosing the appropriate operating range is just as important as accuracy; in fact, the two are directly related. Accuracy is usually expressed as a percentage of full scale or a measurement ratio. The cost of operating under extreme conditions is a sacrifice of accuracy, so it's important to clearly define the manufacturer's definition of the optimal operating range. The capacity toolbar shows the optimal operating range. The PID toolbar illustrates a simple flowchart and the equipment it supports. We can further refine the headings, considering the installation and purpose of each device. Once the purpose and functional parameters of each device are determined, the next step is to choose a specific manufacturer and product model. How to Express Accuracy When a device is used for measurement, it provides an approximation because no device can be absolutely precise. There will always be a difference between the measured value and the ideal value caused by some factor. In practical applications, the engineer's goal is to reduce this difference so that it can be ignored in the process context. The degree of difference between the measured value and the ideal value is the accuracy class. The area between the upper and lower tolerance limits is the error range. According to the Institute of Instrumentation, Systems and Automation (ISA), there are five ways to express the accuracy class: 1. Expressed in terms of the measured variable itself. For example, 100psi ± 2psi. 2. A percentage of the measurement range or scale. 0-500psi pressure is a ±1% scale, with an error range of ±5psi at any point within the entire measurement range. 3. A percentage of the upper limit of the measurement range. (This value is the same as the second value when the lower limit of the measurement range is 0.) 4. A percentage of the measurement span, used when the lower limit of the measurement is not 0, for example: an accuracy class of 100-200℃ for a temperature sensor. 5. Percentage of Actual Measured Value: The 0-100 psi pressure rating is 1% of the actual measured value. (This means that the accuracy is ±0.1 when the measured value is 10, but ±1 when the measured value is 100. This indicates that lower measured values ​​have higher accuracy, but this is less common.) Most equipment uses methods 1, 2, 3, 4, or a combination of these methods, linking accuracy to the measurement range or upper limit. In fact, for all types of equipment, the measurement range is related to accuracy in some way. Example of Process Instrumentation This simple chemical mixing flow diagram illustrates the basic principles of equipment selection. Here, water is added to the chemical reagent to obtain the diluted final product. Changes in process conditions require the mixing ratio to vary within a narrow range. All flow and pressure transmitters convert the measured values ​​and transmit them to the PLC; most instruments display the measured values ​​instantly. The reagent flows through a pump, a check valve, and an electro-pneumatic control valve. After passing through an electromagnetic flowmeter, the liquid, to ensure unobstructed flow and high measurement accuracy, also flows through a pressure transmitter before the mixing point. The water first flows through a filter, with a differential pressure sensor connected across the filter to monitor its condition. If the differential reading exceeds a set value, an alarm is triggered, indicating excessive residue at the filter. After passing through a check valve and pump, a control valve regulates the flow rate to maintain the reagent ratio. A pressure transmitter and vortex flowmeter follow the control valve to measure throughput. For this less critical measurement point, a simple and inexpensive vortex sensor suffices. After the two components are mixed, a pressure transmitter monitors the mixed solution before the static stirrer. Before the mixture flows to the storage tank, an orifice plate and differential pressure transmitter measure the final volume. Both feed lines are checked to ensure the accuracy of the final volume. All these factors are linked by a PLC, which then adjusts the control valve to ensure the correct volume and dilution ratio. One scale cannot meet all needs. Since measurement range and accuracy are so closely related, choosing equipment with an appropriate measurement range is just as important as choosing the correct accuracy class. Factory engineers have a tacit rule that the measurement range of pressure or other equipment should be selected according to the following principle, even if the measured value during normal operation is 2/3 of the full scale. Although this statement originates from the era of mechanical and analog equipment, it remains a good suggestion. Equipment performance is rarely perfectly linear. Linearity is used to represent how well an actual measurement curve approximates a straight line. Within the measurement range of a sensor or instrument, some parts have better linearity than others, especially at the lower part of the curve where inaccuracies are severe. Therefore, some manufacturers use the measurement ratio at the upper limit of the measurement range to describe accuracy (measurement ratio = maximum measured value / minimum measured value). For example, accuracy can be expressed as a percentage of the upper limit of the measurement range, such as a 4:1 measurement ratio. This means that the accuracy class is only applicable to the upper 3/4 of the measurement range; the lower 1/4 has a different error rate, generally a higher one. Other types of equipment have their own characteristics and are not suitable for expressing accuracy levels using the methods described above. These devices typically calculate their accuracy level using a series of variables related to sensing technology, sensing devices, and transmitters. Process and environmental factors can also be taken into account. Generally, these devices have very high accuracy, require careful analysis for selection, and are used in specialized applications. Because equipment suppliers strive to package their products well, they often include instruction manuals with their product lines. However, when each supplier uses different technologies, the same descriptions of performance can be given different meanings, or related factors can be separated, leading to many such inconsistencies in the manuals. David W. Spitzer, author of the third edition of *Industrial Fluid Measurement*, stated, "Reviewing technical data for non-contact level gauges from 60 suppliers, I found approximately 30 variables for expressing accuracy." Spitzer says, “We often have to deal with some erroneous omissions because suppliers might not tell you everything, or they can’t tell you everything at the same time. For example, a flow meter might have an accuracy of ±0.5% and a measurement ratio of 1000:1. What the supplier can’t tell you is that these two parameters cannot both be true simultaneously. Further research shows that ±0.5% accuracy only occurs when the measurement ratio is around 25:1, while the accuracy is 4% at a measurement ratio of 1000:1.” How do other factors work? Accuracy can be affected by environmental or process conditions. For example, a flow meter is very accurate at a specific temperature. If the fluid temperature or ambient temperature changes, even if the fluid is the same, the output value may deviate. In more serious cases, this deviation may exceed the equipment’s allowable error range. Sophisticated instruments are often designed to monitor these external influences and automatically correct for key variables. Carefully reviewing the supplier’s technical specifications may provide information on the correction factors for these external influences, or at least that the so-called accuracy is based on certain ideal conditions. If this information is unclear, it is best to ask the supplier. Equipment Selection Once the process requirements, such as functionality, accuracy, and measurement range, are clearly defined, you can begin selecting suppliers and creating lists. Instead of choosing a single vendor, use an elimination process. Robert Mapleston, Marketing Communications Manager at ABB Instrumentation, frequently sees this. He says, “Salespeople are still very important. Customers like to do preliminary research, and you need to carefully publish information so that customers are prepared before they call. Then they'll call and say, ‘I want to use this pressure gauge in my project.’” Equipment quality has improved across all areas and price ranges. This frees the selection process from tedious analyses of accuracy, functionality, and reliability. “The market is very different now than it was 15 years ago,” says Mapleston of ABB. “Back then, it was all about features and performance. Now, the starting point for technology isn’t that different, and customers are looking for different things. Most of them have only a small number of maintenance personnel, so maintenance dictates their choices. How do they want to make their selections? They want equipment that is more robust, easier to maintain, easier to set up, and of course, a lower price.” Scott Saunders, vice president of sales and marketing at Moore Industries, offers a more cautious perspective: “While manufacturing processes and component sourcing are indeed stable, you need to think long-term. Some low-priced sensors and transmitters have started to become inaccurate and unstable; they won’t work forever.” Beyond long-term performance, Saunders also sees differences in protection for less sophisticated instruments. “If a worker uses a walkie-talkie next to a poorly filtered thermocouple transmitter, it can cause radio frequency interference, leading to temperature spikes in the system, tripping alarms, and stopping operation.” Considering the interdependencies between devices and the network of equipment, keeping the entire system in mind is crucial. Natalie Strehlke, Product Manager at Emerson Process Management's Rosemount division, asked, "You have to be clear about how critical this application is, and what you're going to measure?" "As a company, our focus has shifted from the product to the application. Customers don't just want a process variable; they want control over the entire plant. They ask us, 'What else can these devices do?' Customers demand performance, that's true, but they also demand reliability and stability. Reliability means how much time I need to spend on maintenance. Stability means I don't need to calibrate as often." Measurement range versus accuracy: which is more important? Your job is to measure the pressure of a liquid in a tube, say 40 psi. You have two pressure gauges with different ranges: Gauge 1: 0-300 psi, Class 3A (±0.25% of the range) Gauge 2: 0-100 psi, Class 2A (±0.5% of the range) The first gauge is better because it's twice as accurate as the second, right? Yes, that's true, but not in the following situations. Table 1 achieves ±0.75 psi (0.25% of 300) at any point, while Table 2 achieves ±0.50 psi (0.5% of 100) at any point. When utilizing the higher accuracy portion of the range, a lower-accuracy meter will be more accurate than a higher-accuracy meter. The key to selection is to ensure the operating range of the chosen equipment is as close as possible to the actual operating pressure, while also considering protection against instantaneous spikes. As long as you can guarantee that instantaneous spikes do not exceed the instrument's upper limit, a 0-50 psi 2A class meter can provide more accurate measurements. Measurement accuracy itself is only one factor in selection; in this example, the measurement range is equally important. Which characteristics are important? If accuracy is sufficient for most customers, what should be considered among the many alternatives? ■ Functionality – What else can the equipment do besides providing process variables? ■ Ease of programming – Can your operators adapt to the system, or will they require additional training? ■ Network Scalability – This device may meet your needs today, but are there plans to upgrade it to a fieldbus system or DCS system in the future? ■ Self-Diagnostic Functions – Besides providing process variables, what environmental parameters can this device provide? What kind of error messages or self-status monitoring do you need? ■ Stability – How often does this device need calibration? Is your current device mechanical or analog? If so, you may calibrate it frequently. Factory-calibrated equipment can be used directly and operate for many years before the next calibration. This reduces maintenance costs. ■ Reliability – Is maintenance of a device frequent? Is the device's installation location easily accessible for maintenance, or is it difficult to access, or is the installation location dangerous? Is the low initial investment worthwhile for modifying unreliable equipment? It's difficult to obtain substantial information without relying on your own experience. ■ Scalability – Do you need to replace equipment to upgrade your fieldbus and network systems? Many vendors offer modular designs that can be upgraded as needs change. ■ Factory User Experience – What products are you using? Are they performing well? If you have contact with a manufacturer, dealt with their sales staff, or have their products... At least you've made a worthwhile investment. Saunders says, "The best compliment you can give to equipment vendors is to say, 'We installed your sensors and we don't have to worry about them.'"