The design of process cooling systems typically focuses on the equipment that will actually do the cooling. Much is said and done about the type, size and materials of heat exchangers, cooling towers, refrigeration systems, cooling jackets and other cooling equipment.
Once design and selection are complete, consideration should be given to monitoring the flow of process cooling fluid through the equipment. In many installations, failure to monitor these flows reliably can reduce equipment life and increase downtime. In some installations, catastrophic equipment damage can result and produce negative impacts on cost and production as well as affect personnel safety.
Choosing a Process Cooling Flowmeter
Fortunately, cooling systems are relatively forgiving in the sense that the flow of process cooling fluids can safely vary over a relatively wide range of flows. For example, assume that a piece of equipment requires 80 liters per minute (lpm) of cooling water to satisfy its worst-case process conditions, so the equipment will function properly when the process cooling flow rate is 80, 100 or 120 liters per minute or more. However, it will be more expensive to provide the cooling water needed to operate at the higher flow rates.
This range of acceptable flow rates means that the flowmeter does need not be overly accurate. For example, prudently setting the flow rate at 100 lpm would provide a 20 lpm safety margin in the above application to allow the heat exchanger to operate properly even if the flowmeter measures up to 20 percent higher than actual. In this application, a flowmeter that measures within 20 percent of the measured cooling water flow rate might, theoretically, be acceptable. However, installing a flowmeter with better accuracy specified as a percent of the measured cooling water flow might be more pragmatic to ensure sufficient cooling water flow. Be sure to note that these accuracies are percentages of the measured flow — not percentages of full scale or percentages of the flowmeter capacity.
While this example may seem to show that cooling system flowmeters may not need to be that accurate, it may be beneficial for these flowmeters to actually be more accurate than the minimum. Providing more cooling water than is necessary costs money because the water needs to be treated and pumped through the cooling water system. In this application, the pumping cost of providing cooling water flow of 160 lpm to the equipment is twice that of operating at 80 lpm and 60 percent more than operating at 100 lpm. Installing a superior flowmeter that measures flow within 5 percent of the actual flow rate would theoretically allow operation at 85 lpm and further reduce the cost of pumping. Nonetheless, it might be pragmatic to actually operate the equipment somewhere above 90 lpm to provide an additional safety margin for abnormal operating conditions.
While accuracy may be desirable, it is perhaps more important that cooling system flowmeters operate reliably and not fail. It should be clearly stated that many flowmeters have unsafe failure modes in which the flowmeter can measure a seemingly normal flow rate under no-flow conditions. A flowmeter that indicates flow when no flow is present does not provide protection for the equipment and can potentially pose a significant personnel safety hazard. Unsafe failure modes can occur in both electronic and mechanical flowmeters. However, flowmeters that use mechanical principles are typically more susceptible to fail unsafe than are flowmeters based on electronic principles.
Process Cooling Flowmeter Types
There are many flowmeters that can be used in process cooling applications. Perhaps the simplest device is a visual sight flow indicator where the operator can observe the turning motion of the indicating rotor in the flow. This device does not actually measure flow, but higher flow rates will turn the rotor faster. In time, operators will get a sense of when the rotor turns slowly, thereby indicating a potential problem.
Visual sight flow indications are subjective in that different operators will interpret the rotor speed differently. As a result, the actual process cooling flow rates are typically set significantly higher than necessary to ensure that all of the operators recognize low flow conditions before the flow is actually too low. Using the example above, the operators will likely not be able to determine if the flow falls below 80 lpm when normal operation is at 85, 90 or 100 lpm. The system may have to be operated at (say) 160 lpm in order for all of the operators to reliably detect low flow.
Glass-tubed and plastic-tubed float variable-area flowmeters are similar to visual sight indictors in that the fluid is visible to the operator. Float variable-area flowmeters have a float that rises when flow increases. Vane-style variable-area flowmeters have a vane that increases its rotation when the flow is increased. The operator can read the flow rate by using the float height or vane position in conjunction with graduations on the flowmeter. Variable-area flowmeters are commonly used on small pipes under 1 to 2".
Variable-area flowmeter accuracy is usually specified as a percentage of full scale, so be careful determining the desired flow rate to ensure adequate process cooling. Using the continuing example, a variable-area flowmeter with a full-scale flow of 200 lpm and an accuracy of 5 percent of full scale would have an absolute accuracy of ±10 lpm. Setting the process cooling flow rate at 90 lpm would appear to be reasonable to ensure adequate cooling. However, a higher setting should be used to account for other factors than can affect the measurement such dirt and coating.
In a bypass arrangement, a differential pressure flowmeter element is in the main flow of a larger pipe main and a variable-area flowmeter. The differential pressure element passes the overwhelming majority of the flow. Calculations are performed to determine the relationship between the flow through the bypass and the total flow so that the total flow rate can be inferred from bypass flow measurement.
Vortex-shedding flowmeters utilize a bluff body in the flow stream to create vortices that are similar to the vortices that are created downstream of a rock at the surface of a flowing river. The frequency of vortex generation is proportional to the fluid velocity — with the caveat that no vortices are formed at low flow rates. Vortex-shedding flowmeters in process cooling applications usually can be sized to operate high in their measurement range so this constraint is typically not an issue. In addition, their relatively robust construction with no moving parts makes them rugged and reliable.
Many of the flowmeters described are relatively simple devices, but their operation can be labor intensive, energy intensive and prone to subjective results. Further, many are only active when they are actually observed by the operator. At best, observations may be performed hourly for only a few seconds. At worst, observations are made sporadically, perhaps once a month or less.
Variable-area (vane and float) and differential pressure flowmeters are available with transmitters that can provide continuous process cooling flow measurements for monitoring and alarming process cooling flows. Vortex-shedding flowmeters are electronic and can inherently provide continuous measurements.
Flowmeters with transmitters provide objective measurements that are not subject to varying degrees of operator discretion, and they can provide continuous protection, which enables the operator to focus on more productive work. For example, an operator may actually look at a local process cooling indicator for a few minutes (or seconds) during an 8-hr shift. This relatively low coverage increases the chance that an operator will miss a transient low flow event or perhaps not notice an event before equipment damage occurs. By contrast, a control system can monitor and alarm a more sophisticated process cooling flowmeter with a continuous measurement signal and detect all events that occur during the 8-hr shift when they occur. As a result, the operator is better able to address the event immediately, before the event affects the process. In addition, the operator can focus on other important activities with the time saved by not having to walk around to observe the local indicators.
Process cooling flowmeters are an important part of plant operation because they protect expensive plant equipment and plant personnel. Without these seemingly mundane instruments, much of the plant equipment could be in jeopardy if it is not operating safely and efficiently. The choice of flowmeters available for process cooling service is large, but visual sight flow indicators, variable- area flowmeters (vane and float), differential pressure flowmeters and vortex-shedding flowmeters represent pragmatic selections to ensure that process cooling flows are sufficient to protect equipment and personnel.