DAVID W SPITZER'S E-ZINE (July 2023)

The following equation expresses this relationship when the transmitting and receiving devices are flush with the pipe wall.

It should be noted that some suppliers simplify this equation by substituting the square of the transit time for (Tu * Td) in the denominator.

In some designs, sensors may not be mounted flush with the inside pipe wall, so part of the ultrasonic energy path may travel where there is no net flow. Because T_u and T_d appear as a product in this equation, the measured transit times should be corrected to account for the time to travel through the no flow distance.

Excerpted from The Consumer Guide to Ultrasonic and Correlation Flowmeters.

Replacing the orifice plates was one of the recommendations made in a report auditing a number of orifice plate custody transfer installations. Manufacturing the orifice plate would seem to be straightforward because it can be fabricated using specifications in existing standards that are available from a number of sources.

Orifice plate standards have smoothness requirements that result in an orifice plate that is relatively smooth to ensure that the appropriate geometry of this physical aspect of the orifice plate is maintained. On the other hand, the user had an internal requirement that the equipment located in the piping requires relatively rough surfaces. Investigation revealed that the source of this requirement was the user's experience that equipment with relatively smooth surfaces tended to leak over time. Therefore, plant-wide specifications required rough surfaces to avoid leakage by the gasket.

The orifice plate supplier was "pulling his hair out" regarding the smoothness of the orifice plate surface because he had to deal with these conflicting requirements in the same orifice plate. In particular, standards require a smooth surface to achieve accurate measurement while the plant (his customer) required a rough surface to avoid leakage.

The seemingly conflicting requirements were resolved after some initially frantic discussions. Once the facts were determined, it became clear that the orifice plate needed to be smooth for accurate measurement. Yet it also had to be rough to properly seal the piping. But the measurement occurs in the flowing fluid whereas the sealing occurs away from the flowing fluid outside of the pipe per se.

It is important to recognize that the smooth and rough requirements apply to different parts of the orifice plate even though neither requirement explicitly states this. Manufacturing can then proceed by fabricating the entire orifice plate per the relatively smooth orifice plate specifications. Afterwards, the part of the orifice plate that is not used for measurement (the non-wetted annular area located outside of the inside diameter of the pipe) can be made rough so as to meet the sealing requirement of the gasket.

It took a little convincing... but the orifice plate was eventually fabricated to meet both seemingly conflicting requirements.

This article originally appeared in P. I. Process Instrumentation magazine. 

1. 0.50 percent of rate
2. 1.00 percent of rate
3. 1.01 percent of rate
4. 2.00 percent of rate
5. None of the above

The flowmeter portion of a flow measurement system can be thought of as consisting of a primary element, transmitter, and a signal converter. The primary element is generally in contact with the flowing fluid (but may not necessarily be wetted). The transmitter processes the signal from the primary element to produce the flow measurement. The signal converter changes the flow measurement into an instrument signal (such as 4-20 mA) for transmission to another device.

The stated accuracy of the flowmeter (1% of rate) typically refers to the accuracy of the primary element and transmitter portions of the flow measurement system. Examination of the flowmeter specifications will likely reveal that there is a signal converter accuracy that should be added to the stated accuracy to reflect the accuracy of the analog output. The signal converter accuracy is typically expressed as a percentage of full scale, so its value expressed as a percentage of flow rate will be different at different flow rates. Therefore, Answers A, B, C and D are not correct and Answer E is correct. Further, the accuracy of the control system input board should not be considered in calculating the accuracy of the analog output.

Additional Complicating Factors

The overall flow measurement system error should include all the components in the system. As such, it should include the accuracy of the control system input board and subsequent calculations. It should be noted that while some system inputs can exhibit accuracies on the order of 0.01 percent of full scale, other system inputs can exhibit accuracies as high as 0.4 percent of full scale. These poorer accuracies can profoundly degrade measurement accuracy when the flow is low in the flow range.

This article originally appeared in P. I. Process Instrumentation magazine.