Flow meter accuracy specifications can be confusing. Here is how to interpret them correctly and understand what they mean for your application.
Accuracy vs. Repeatability
Accuracy describes how close the reading is to the true value. Repeatability describes how consistent readings are under the same conditions.
A meter can have good repeatability but poor accuracy, meaning it is consistently wrong by the same amount. For custody transfer and batching applications, you typically need both good accuracy and good repeatability.
Types of Accuracy Specifications
Percent of Rate
An accuracy of ±0.5% of rate means the error is proportional to the actual flow. At 100 GPM the error is ±0.5 GPM. At 10 GPM the error is only ±0.05 GPM. This provides the best accuracy across the entire measurement range and is how Coriolis and mag meters are typically specified.
Percent of Full Scale
An accuracy of ±0.5% of full scale (where full scale is 100 GPM) means the error is constant regardless of the actual flow. At 100 GPM the error is ±0.5 GPM, which is 0.5%. But at 10 GPM the error is still ±0.5 GPM, which is now 5% of the reading. This means accuracy deteriorates significantly at low flows.
Combined Specification
Some meters are specified as ±0.5% of rate plus 0.1% of full scale. This accounts for both error sources and is a more honest representation of real performance. It is common with quality instruments.
Turndown Ratio
The turndown ratio describes the range over which the stated accuracy is maintained. A 10:1 turndown means the meter is accurate from 10% to 100% of its maximum flow. A 100:1 turndown means accuracy from 1% to 100% of maximum flow.
Typical turndown ratios vary by technology. Coriolis meters achieve 100:1 or better. Mag meters also achieve 100:1 or better. Vortex meters typically achieve 20:1 to 40:1. Turbine meters typically achieve 10:1 to 20:1.
Reading Manufacturer Datasheets
When reviewing datasheets, pay attention to the reference conditions under which accuracy was determined, including temperature, pressure, and viscosity. Check whether the meter was factory calibrated or field calibrated. Look for whether the specification uses accuracy or uncertainty, as these have different statistical meanings. Verify the flow range over which the stated accuracy applies.
Real-World vs. Laboratory Accuracy
Laboratory accuracy is achieved under ideal conditions with perfect installation, stable temperature, known fluid properties, and fresh calibration.
Real-world accuracy is typically 2 to 3 times worse due to installation effects, temperature variations, aging and drift, and differences between actual and assumed fluid properties.
As a practical rule of thumb, expect real-world performance to be 2 to 3 times the published accuracy specification. If a meter is specified at ±0.5%, expect ±1.0% to ±1.5% in practice unless installation and conditions are carefully controlled.