The ultrasonic energy travels from the sensor to the material and back. The energy can be attenuated in transit due a number of phenomena, including acoustical attenuation caused by nature of the vapors in the energy path, and dust/dirt in the energy path to/from the material. Note that ultrasonic level measurement can often be performed accurately and reliably even when the material cannot be seen visually. The presence of dust/dirt in the energy path can be temporary, such as when filling occurs and causes a dust cloud to form in the energy path.

The material itself can cause the intensity of the reflected ultrasonic energy to degrade when the material exhibits poor reflective qualities, such as when contaminants on the surface of the material cause the ultrasonic energy to reflect poorly.

In addition, accuracy can be degraded based upon the surface on which the ultrasonic energy is reflected. For example, ultrasonic level measurements typically measure the top of a layer of foam by reflecting off the top of the foam. However, the characteristics of the foam, such as its composition, density, bubble size, and the like, can cause the foam to absorb ultrasonic energy instead of reflecting it. In this application, not only will the ultrasonic level measurement system not measure the (desired) liquid level, but varying amounts and consistency of foam can cause the measurement to become erratic.

Excerpted from The Consumer Guide to Non-Contact Level Gauges.

Repeatability is the ability of a flowmeter to reproduce a measurement each time a set of conditions is repeated. Flow measurements taken using a flowmeter exhibiting poor repeatability would be chaotic. For example, if measurements were taken of a known flow rate of 100 units per minute, a flowmeter with poor repeatability might measure 85, 101 and 93 units per minute on three consecutive days. If these measurements were used to feed material to a process at a given flow rate, different amounts of material would be fed to the process at each of these times. The operator would be at a loss to determine what the flow setting should be to obtain a flow of 100 units per minute. This amount of variation could be detrimental to the operation of the plant.

In a similar test, a flowmeter with better repeatability might measure 96, 94 and 95 units per minute. Note that the difference between the measurements is smaller, that is, the measurement is more repeatable. From experience, the operator will find that setting the flow rate at 95 units per minute results in the desired plant operation of 100 units per minute. As such, one could make an argument that to operate the plant in a steady manner it is desirable to use flowmeters that are repeatable. Note that the flowmeter setting does not correspond to the desired flow.

Accuracy is the ability of the flowmeter to produce an output that corresponds to the characteristic curve of the flowmeter. Note that a flowmeter that is not repeatable cannot be accurate. Stated differently, if the output of the flowmeter is chaotic, it cannot correspond closely with the characteristic curve. Therefore, in order for a flowmeter to be accurate, it must be repeatable.

In a test similar to that described above, an accurate flowmeter might measure 101, 99, and 100 units per minute, and the operator will find that setting the flow rate at 100 units per minute results in the desired plant operation of 100 units per minute. Note that the flowmeter setting does correspond to the desired flow.

Many applications will be able to function with a flowmeter that is repeatable. If this is all that is necessary, maybe a repeatable flowmeter should be installed. For example, if a tank level increases, its level controller will increase its effluent flow setting to maintain the level setting. In this example, it is seemingly not important to accurately measure the flow rate, but it would be beneficial if the flowmeter were repeatable so that equal flow setting changes result in equal flow changes. Repeatable flowmeters are often less expensive than accurate flowmeters, providing another incentive for their use.

However, often overlooked is the opportunity and sometime necessity of performing process calculations in order to ensure proper economic process operation or improve the process. For example, if the effluent flowmeter cited above is the overhead take-off of a distillation column, the flow measurement may be used to calculate the column reflux flow rate. The original plant design may not have contemplated this, but operating experience may have determined that this control strategy will provide better control of the column. Many such applications exist, and many of these are "discovered" after the plant is operating. Sometimes, repeatable flowmeters must be replaced with accurate flowmeters to obtain operating data and/or accommodate these improvements.

 Accurate flowmeters are desirable because they are repeatable and yield measurements that closely reflect the true flow rate. Repeatable flowmeters may not yield accurate measurements, but they will perform in the same manner under the same conditions. Which is appropriate depends in part on the application and budget. 

This article originally appeared in Flow Control magazine. 

This problem is similar to one that appeared in the second part of the Professional Engineering examination for control systems engineering. With the information given, it would be academic to calculate the differential pressures for each flowing condition and select the correct answer. However, there were 40 questions to answer during the 4-hour examination (or 6 minutes per question), and it quickly became apparent that the time required for the calculations would greatly exceed 6 minutes.  My approach to the problem was pragmatic --- either skip the problem or find a better way to solve it.

Let's get back to the basics. Orifice plate flowmeters typically operate in the turbulent flow regime.  A quick Reynolds number calculation at 100 gpm confirms this with an operating Reynolds number on the order of 180,000. In the turbulent flow regime, the differential pressure across the orifice plate flowmeter is a function of the square of the flow rate through the flowmeter. I remember this relationship with the simple expression, "Double the flow... four times the differential." Conversely, reducing the flow rate in half will reduce the pressure drop to approximately (0.5)², or 25 percent of the original pressure drop.

In this problem, increasing the flow rate from 90 gpm to 125 gpm will increase the differential pressure by a factor of (125/90)² or 1.93. Examination of the answers reveals that only Answer A presents differential pressures that approximate this relationship.

Additional Complicating Factors

The amount of differential pressure produced is also affected by fluid density. At a given flow rate, for every 1 percent that the fluid density increases, the differential pressure will be reduced by approximately 1/2 percent.  In liquid applications (such as above), fluid density does not vary much, so its effect is relatively small.  In gas applications, changing operating pressures (and sometimes temperatures) at the different flow rates can significantly affect fluid density, and cause the differential pressure developed across an orifice plate flowmeter to not follow its quadratic relationship as a function of flow.  

This article originally appeared in Flow Control magazine.