Throughout industry, flowmeter accuracy and repeatability are scrutinized with different criteria to ensure confidence in a particular application.
Q: How do you know if the measured value is correct?
A: The confidence of measured value agreement resides with the original manufacturing laboratory, the range of uncertainty of the technology being used, and the manufacturer''s practices and accreditations.
Field proving activities can be used as a point of reference, but may be influenced by a number of factors. These factors include, but are not limited to, the accuracy or repeatability of the comparative standard used, calibration interval of the master standard, comparison procedure, unit conversion error, failure to perform field adjustments at reference conditions, failure to consider all error factors, or actual test meter drift or error outside of manufacturing tolerances.
Consequently, the best recommendation is to consult with the manufacturer and ask for help or a re-certification at the calibration facility.
Q: What options exist in calibration related to flowmeters?
A: Flow calibrations are performed at flowrates typical for the meter size and based on standard measurement points defined from a history of like meter calibrations.
This does not mean a custom calibration could not be performed at unique points like low flow ranges. In fact, special calibration points and ranges are often specified on meters for critical applications. As more points are chosen, the cost required to perform custom calibrations becomes more expensive. Stipulations do exist, however, and the manufacturer should be consulted prior to specification of custom calibration. For example, Coriolis mass flowmeters are only calibrated in mass flow units.
Q: Why was my meter not tested at the full-scale flowrate?
A: Meter uncertainty (total error) is greater in low flow ranges for all flow technologies. A statement of flow meter total uncertainty usually includes: percentage of rate plus a percentage of full scale.
Tests on thousands of flowmeters have shown that to meet the published accuracy, a series of tests are run at low and partial full-scale values to ensure they will meet their uncertainty tolerance when operated at full scale. Generally, the meter is calibrated with flowrates of 5 percent and 20 percent of their full-scale capabilities since linearity increases with an increase in flowrate, thus reducing customer costs by reducing high calibration flowrates and the associated manufacturing costs.
Q: How do I interpret the accuracy of the calibration performed at various points documented on the calibration data sheet?
A: The difference between the target (the value measured by devices in the calibration laboratory) and the measurement (the value from the flowmeter being calibrated) divided by the flowrate used for a specific calibration point equals the error as a percentage of the actual flowrate. For a calibration to be successfully completed, the documented points must all fall within the published accuracy specification of the flowmeter. Dots define the calibration points on the calibration data sheet on a graph, which must fall within the boundary lines that indicate allowable error.
Q: What are the options for flow calibration of my flowmeters made by a traceable calibration laboratory?
A: Most flowmeters undergo a traceable flow calibration prior to shipment to ensure accuracy and repeatability. A hardcopy calibration data sheet is produced for each meter. The test points, measurement uncertainty determined, and identification of the meter (i.e. serial number, model number, and test person and date) are noted.
The calibration data sheet can typically certify a minimum of two, up to 10 measurement points against a gravimetric standard. A tolerance band of acceptable error is shown on the calibration sheet within which all test points must fall. The error limits are defined for each meter series and type as defined by the manufacturer. A repeatability test is conducted during the calibration to ensure meter confidence.
Q: What is the difference between an ISO-17025 accredited calibration facility and those who claim they are NIST traceable?
A: There are several production facilities that conform to the National Bureau of Standards (NBS) or are self-certified as traceable to NIST, the United States National Institutes of Standards and Technologies.
Unfortunately, NIST does not govern calibration facility claims of traceability. The responsibility of NIST-traceable claims reside with the individual laboratory and in some circumstances may not produce acceptable measurement results. Consequently, NIST traceability is not acceptable in all foreign countries.
A more rigorous third-party, ISO-17025 accreditation of the calibration facility is preferred. ISO 17025 can be certified independently by The American Association for Laboratory Accreditation (A2LA). Unlike NIST, ISO 17025 ensures methods, practices, a facility’s personnel, administrative systems, documentation, and proficiency testing are traceable to the International System of Units (SI). The SI is the primary standard of reference used by NIST and numerous foreign countries.
The ISO-17025 accreditation procedure evaluates the total measurement uncertainty (accuracy and repeatability) of each component on the calibration rig and is a means of comparing ISO-17025 accredited rigs to each other. Additional inter-laboratory proficiency testing is used to provide individual laboratory confidence.
Q: How can I conform to governmental agency requirements?
A: Governmental conformance is a broad subject to cover without addressing a few of the typical agencies that flow metering data is reported.
Meters used for legal trade or custody transfer governed by the U.S. Department of Commerce conform to requirements adopted by the National Conference of Weights and Measures group and NIST Handbook 44 requirements by section for the technology they represent. NTEP-approved Coriolis mass and electromagnetic flowmeters, for example, can be used for legal trade in various industries.
Throughout industry, flowmeter accuracy and repeatability are scrutinized with different criteria to ensure confidence in a particular application. In utility environments, for example, a government body may require a flowmeter to meet minimum accuracy and repeatability standards. In others, say private industry for example, in-house specifications are likely the source for accuracy and repeatability standards. Whatever the case, be it a government-mandated standard or an internal performance requirement, flowmeter accuracy and repeatability are of the utmost importance.
There are a variety of methods users can employ in an effort to ensure flowmeter performance, but the terminology for describing such practices is oft used interchangeably despite the unique nature of different practices.
In some applications, flowmeter verification will satisfy requirements for meter performance within a defined tolerance of the original manufactured state. In other applications, however, a traceable calibration is required to fulfill this requirement. And some other scenarios may require in-field proving of the meter to instill confidence in its performance in a specific system.
These techniques, which are the most prominent methods for ensuring flowmeter accuracy and repeatability, have inherent differences and can be a good fit in many environments, depending on the requirements of the application under consideration.
Calibration
There are several design standards for calibration systems. These may include, but are not limited to, volumetric methods, gravimetric methods, and master meter comparison. These methods can produce results with an uncertainty of better than 4-to-1 as compared with the meter to be tested.
In addition, there are standards that govern and encompass the entire calibration system. NIST and ISO 17025 define standards and requirements with varying degrees of complexity for calibration facilities and procedures. The National Institute of Standards and Technologies, or NIST, defines standards for traceability that reside with the individual laboratory for maintenance or self-compliance. ISO 17025 is a more rigorous, third-party accreditation. This standard encompasses the entire calibration system and produces metrics for the calibration rig components, administrative systems for process operations, personnel proficiency, and documentation supporting the traceability and total measurement uncertainty for the entire calibration facility. ISO 17025 standards ensure the highest level of confidence in accuracy and repeatability.
The term calibration is reserved for manufacturers that use traceable standards to establish or correct factors specific to an individual meter. As part of the calibration process, the meter to be tested is compared with the laboratory master standard. ISO 17025-certified meters can achieve +/- 0.05 percent accuracy.
The calibration process can take place under two basic scenarios, i.e., as last step in the original manufacturing process or for re-calibration. In simple terms, the purpose of calibration is to determine whether deviations are present due to manufacturing or whether they have occurred from process usage.
In the first instance, the device output signal is adjusted to match the target value of the calibration system. Although special calibration protocols exist, the process is generally a two-point process. The goal is to check and establish the zero point stability when appropriate. Once established, further points within the range are used to check the linearity of the meter to be tested. These points are often repeated to ensure the proper adjustment according to observed performance. Uncertainty can then be identified and accounted for to ensure that a linear and repeatable flowmeter can realize final delivery.
In a re-calibration scenario, an operational flowmeter is returned to a laboratory for the purpose of creating up-to-date documentation to ensure meter conformance to a traceable standard or to rectify mechanical and electronic deviations in the meter as a result of sensor aging or component drift.
In-Field Proving
In-field proving can be done in-situ in a customer application or out of process with a flow proving cart, master meter, transfer standard, or scale. Meter proving allows a master meter to be piped in series with the device under test. Care must be taken to limit partially filled pipes, pressure losses, velocity effects, or flow profile influence when installing a master meter in the process piping. The master meter can be a meter with the same or higher accuracy to confirm agreement with the meter to be tested. For example, measuring uncertainties of +/- 0.1% of rate are achievable with modern flowmeters.
The level of accuracy for a proving meter should match the customer’s expectations for the proving process. Testing runs can be made at continuous flow or totalized directly for comparison. Unlike the controlled conditions of a calibration laboratory, field-proving competency resides with the owner of the master meter, employed practices, and ability to reproduce test conditions over the lifecycle of the process meter.
Flow proving is sometimes used as a method to modify a flowmeter’s calibration factor to match the master meter standard. Field proving may provide a justification for removal and return of the process meter to a flow calibration facility or laboratory for an “as found” calibration to rectify the deviations and re-establish the output factors in accordance with the manufacturer’s specified tolerances.
Flow proving cart component meters, master meters, or transfer standards are usually sent back to the respective manufacturer or appropriate authority for traceable calibrations to ensure they provide a proper metric for on-site proving. Installation guidelines for the proving meter/standard are vital to ensure the proving standard traceability is transferable to the field installation it is intended to prove. It is also important that all electrical area classifications are considered when designing and applying flow proving carts, master meters, or transfer standards. There are often multiple classifications within a plant that should be observed.
Using a nonintrusive ultrasonic flowmeter is an ideal situation in many applications. This practice should not be generally confused with proving as described earlier in the article. Ultrasonic flowmeters are generally able to produce results that are at the 2-5 percent of rate accuracy level. This is typically less accurate than themeter to be tested. It can, however, be used to provide a sense of confidence to verify that the meter to be tested is performing satisfactorily within an expected range.
The two latter methods — electronic verification and nonintrusive comparison — are quite practical and economical in the correct applications. Not only do they allow for simple checks to ensure the primary measuring device is functioning satisfactorily, they are portable and allow for repeated use across a broad range of installations. Limitations on measuring accuracy and the suitability of data must be taken into account.
Flowmeter verification is often used as a confirmation of long-term sensor or transmitter stability. Verifications can be electronic simulations that do not involve actual flow or process comparisons under flowing conditions.
Electronic verification does not include a wet test of the process meter. The electronic verification device coupled with integral meter diagnostics can detect changes in the sensor geometry from coating, erosion, or corrosion for example. In general, however, the electronic verification is used to investigate electronic component drift, transmitter or converter errors, or hardware failure. These tests can be confirmed or compared with the original manufacturer tolerances and specifications over the process meter lifecycle.
Verification devices can store past tests so they can be reviewed or compared over the flowmeter lifecycle. These tests can be stored as read-only electronic files or printed for hardcopy review. As an option, electronic flowmeter verification can be provided at the time of manufacturing to create a baseline analysis from which to judge the long-term changes that occur in a tested sensor or transmitter component. Repeat field verification tests can show long-term change in meter stability. In some cases, verification tests may be used as a means to satisfy quality systems, government, or agency requirements. A field verification activity can be carried out in-situ or offline.
Electronic verification tools should be periodically returned to the original manufacturer or laboratory to ensure proper functionality and calibration of electrical components.
Another method for verifying flowmeter operability in-situ is by comparing an additional flowmeter or flowmeter technology. This can be an insertion meter, such as a hot-tap differential pressure, magnetic, or ultrasonic device. In many cases, however, a nonintrusive ultrasonic flowmeter is a better fit because no process interruption is necessary. Nonintrusive devices can be used when pipe diameter, pipe wall thickness, and sound velocity characteristics are known. Typically, the manufacturer produces coefficients and values that relate to the pipe and fluid types for meter commissioning. For example, the speed of sound in a liquid to be evaluated for flow comparison is required for ultrasonic commissioning, but is not necessary for magnetic flowmetering. In addition, the fluid characteristics to be tested should remain somewhat constant during the time of comparison. These characteristics include, for example, fluid density, concentration of chemical solutions, and temperature.
The Importance of Performance
Flowmeter performance is definitely critical for all plant operations. The degree of application criticality must, however, be decided at the plant level to ensure the proper maintenance program is applied for all flowmetering devices.
In an optimum situation, verification, calibration, and proving can all be used in harmony to ensure the desired tolerances of the entire installed base are maintained. But it can be difficult to employ an optimum program for all applications. Therefore it is important for leading plant personnel to hold ongoing discussion to determine the appropriate maintenance program.
These discussions should determine if the meter is used for quality control, regulated production control, plant operations, or utilities. Proper planning should take place to procure the appropriate methods, logistics, and tolerances for meter maintenance. For example, planners should determine what meters must be calibrated by the original manufacturer to ensure the highest level of certainty in the device and what devices are candidates for electronic verification by the end-user. Simply put, the user must determine what devices require 1.0 percent of rate accuracy or better and what devices are acceptable at 2-5 percent of rate accuracy.
Taking full advantage of the available technology for verification, calibration, and proving should produce a mix of services and maintenance for the installed base. Properly segmented, these services can result in increased production time, improved maintenance planning, reduction in spare meter inventory, and less costs associated with meter ownership.
About the Author
Jerry Stevens is the business manager of Endress+Hauser’s Flow Products Division. He has 24 years of experience in the specification, sales, and technical support of flow products. Mr. Stevens attended IUPUI in Indianapolis where he studied Mechanical Engineering and Business Marketing. Jason Pennington has worked in application engineering, business development for liquid analytical products, and most recently as a product manager for flowmetering technology at Endress+Hauser. He studied Electrical Engineering at Purdue University''s Indianapolis Campus. Mr. Pennington can be reached at [email protected] or 888-ENDRESS.