In the oil and gas industry, meter proving is one of the most important tools we have for maintaining confidence in liquid measurement. Whether we're measuring crude oil, refined products, or natural gas liquids (NGLs), proving provides a means of evaluating meter performance against a known reference volume.
But there's a misconception that occasionally finds its way into the field:
If a meter proves within tolerance, the measurement must be accurate.
Unfortunately, that's not always the case.
A meter can produce an excellent proving report, demonstrate outstanding repeatability, and still contribute to significant measurement inaccuracies.
At Upland Measurement Services, we believe understanding the difference between repeatability and accuracy is critical to maintaining measurement integrity.
Repeatability vs. Accuracy: What's the Difference?
Two terms that are frequently used interchangeably in the field are repeatability and accuracy. While related, they represent two different aspects of measurement performance.
Repeatability describes how closely repeated measurements agree with one another under the same operating conditions.
Accuracy describes how closely a measurement agrees with the actual or accepted reference value.
Consider a meter that produces the following factors during five consecutive proving runs:
RunMeter Factor11.002321.002431.002341.002451.0023
These results demonstrate excellent repeatability. The meter is producing nearly identical results from one run to the next.
But what does that tell us about the overall accuracy of the measurement system?
Not nearly as much as some might think.
A meter factor of 1.0023 does not inherently indicate an inaccurate meter. In fact, applying a properly established meter factor is precisely how many liquid measurement systems correct indicated volumes.
The concern arises when errors exist elsewhere in the proving or measurement process.
A meter can be extremely repeatable while the system used to evaluate it contains a consistent error.
And consistent errors can be particularly dangerous because they may remain unnoticed for extended periods.
1. A Repeatable Error Is Still an Error
Imagine a proving system where a temperature transmitter consistently reads 2°F higher than the actual fluid temperature.
If that error remains consistent throughout the proving process, the resulting meter factors may demonstrate excellent repeatability.
Every run could agree with the previous one.
The proving report could appear completely normal.
However, depending on the fluid properties, temperature compensation method, and location of the erroneous measurement, the calculated reference volume or corrected meter volume could be biased.
This is an example of a systematic error.
Unlike random errors, which tend to introduce variation between measurements, systematic errors can repeatedly influence results in the same direction.
Common sources of systematic error include:
Incorrect temperature or pressure transmitter calibration
Improperly configured flow computer parameters
Incorrect density inputs or fluid properties
Errors in prover calibration data or correction factors
Inaccurate base conditions
Improper application of meter factors
Instrumentation installed in locations that do not represent actual process conditions
The challenge is that many of these errors may not significantly affect proving repeatability.
Repeatability tells us how consistently the system performs. It does not independently validate every input used to calculate the result.
2. The Prover Is Only One Part of the Equation
A prover is a reference standard used to determine meter performance, but a successful proving depends on much more than the prover itself.
The integrity of the entire proving system matters.
Consider the instrumentation and calculations involved in a typical liquid meter proving operation:
Prover temperature measurement
Meter temperature measurement
Prover pressure measurement
Meter pressure measurement
Prover volume
Fluid density and applicable correction factors
Pulse collection and interpolation, where applicable
Flow computer configuration and calculation methods
Each component contributes to the final calculated meter factor.
An incorrectly configured transmitter or an outdated calibration value can affect the proving result even when the mechanical operation of the prover is flawless.
For example, if the prover and meter temperature measurements are not representative of their respective locations, the applicable volume corrections may introduce error.
Likewise, an incorrect prover base volume or improperly applied pressure correction can bias the reference volume.
The result?
A beautifully repeatable meter factor that may not accurately represent the meter's actual performance.
This is why proper proving requires more than connecting hoses, circulating product, and collecting enough runs to meet repeatability requirements.
It requires an understanding of the entire measurement process.
3. Coriolis Meters: Consistency Doesn't Guarantee Correct Measurement
Coriolis meters have become increasingly common throughout the oil and gas industry due to their ability to directly measure mass flow and provide density measurements.
However, these instruments are not immune to measurement errors.
One particularly important consideration is meter zero.
A Coriolis meter with an incorrect zero offset may produce consistent readings while introducing a measurable bias, particularly at lower flow rates.
Consider a hypothetical meter with a zero-related mass flow offset equivalent to 50 pounds per hour.
At a flow rate of 100,000 pounds per hour, that offset represents approximately 0.05% of the indicated flow.
At 10,000 pounds per hour, the same offset represents approximately 0.50%.
The absolute error remains the same, but its significance increases dramatically as flow decreases.
This is one reason proper zero verification, operating conditions, and flow-rate considerations are important when evaluating Coriolis meter performance.
Other factors that can influence Coriolis measurement include:
Entrained gas or two-phase flow
Improper meter installation or mechanical stress
Process vibration
Changes in fluid properties
Operating outside the meter's recommended performance range
Incorrect configuration or compensation settings
A successful proving under one set of conditions does not necessarily establish the meter's performance across every possible operating condition.
4. What Happens When Operating Conditions Change?
Another commonly overlooked issue is the relationship between proving conditions and actual operating conditions.
Suppose a meter is proved at 400 barrels per hour and demonstrates excellent repeatability.
Following the proving, normal production declines and the meter operates at 100 barrels per hour.
Can we automatically assume the meter maintains identical performance?
Not necessarily.
Depending on meter technology and application, meter performance can be influenced by flow rate, viscosity, fluid properties, and operating conditions.
Positive displacement and turbine meters, for example, may demonstrate different performance characteristics throughout their operating ranges.
Coriolis meters may experience increased relative uncertainty at low flow rates as zero stability becomes a larger contributor.
A meter factor established at one operating point may not accurately characterize performance at another.
This is why understanding the meter's operating envelope is essential.
The objective should not simply be to obtain a passing proving result. It should be to establish a representative meter factor under appropriate operating conditions.
5. The Financial Impact of Small Measurement Errors
One of the most concerning aspects of systematic measurement errors is how quickly relatively small discrepancies can accumulate.
Consider a crude oil measurement facility operating under the following conditions:
Daily Throughput: 10,000BBL
Crude Oil Value: $65.00
Measurement Bias: 0.25%
Daily Volume Discrepancy: 25 Barrels
Annual Volume Discrepancy: 9,125 Barrels
Annual Measured Value Discrepancy: $593,125
A measurement bias of just one-quarter of one percent represents nearly $600,000 in annual measured value at this facility, assuming constant throughput and pricing.
This does not automatically mean an operator is losing that amount of money. The financial impact depends on the direction of the error, contractual arrangements, allocation methodology, and whether the discrepancy is identified and reconciled.
However, the example demonstrates an important point.
Small measurement errors can create substantial financial exposure when applied to large volumes over extended periods.
A meter that consistently proves within repeatability requirements is not necessarily protecting an operator from these types of discrepancies.
6. Looking Beyond the Proving Report
A proving report provides valuable information, but it should not be treated as the sole indicator of measurement system health.
A comprehensive approach to measurement assurance should also consider:
Meter factor history
Are meter factors trending consistently? Have there been unexplained shifts? Do changes correlate with maintenance, operating conditions, or instrumentation adjustments?
Instrumentation verification
Are temperature, pressure, and density measurements accurate and representative? Have calibration records been reviewed?
Flow computer configuration
Are correction factors, base conditions, meter factors, and calculation methods properly configured and applied?
Operating conditions
Was the meter proved under conditions representative of normal operation? Is the meter operating within its intended flow range?
Mechanical condition
Could wear, deposits, entrained gas, valve leakage, or other mechanical issues be affecting measurement performance?
Measurement reconciliation
Do measured volumes reasonably agree with related measurement points, inventory changes, and expected production?
No single indicator tells the entire story.
A successful proving is an important piece of the measurement assurance process, but it is not a substitute for evaluating the complete system.
7. The Difference Between Proving a Meter and Understanding a Measurement System
There is an important distinction between performing a meter proving and evaluating measurement integrity.
A technician can follow the appropriate proving procedure, obtain acceptable repeatability, calculate a meter factor, and generate a report.
But identifying potential measurement problems requires additional knowledge.
It requires understanding how temperature, pressure, density, meter technology, fluid properties, and calculation methods interact.
It also requires recognizing when something doesn't make sense.
Why did the meter factor change after a transmitter replacement?
Why does a meter demonstrate excellent repeatability but consistently disagree with another measurement point?
Why does a Coriolis meter perform differently at low flow rates?
Why are allocation discrepancies increasing even though every meter is being proved on schedule?
These are the types of questions that should drive a deeper investigation.
At Upland Measurement Services, we believe the value of a measurement technician extends beyond the ability to perform a successful proving.
It includes the ability to recognize potential issues, ask the right questions, and help operators understand what their measurement systems are actually telling them.
Final Thoughts: A Passing Proving Isn't the End of the Conversation
Meter proving remains an essential component of accurate and reliable hydrocarbon measurement.
However, excellent repeatability should never be confused with a guarantee of overall measurement accuracy.
A meter can prove consistently while systematic errors elsewhere in the measurement system continue to influence reported volumes.
The most effective measurement programs recognize this distinction and take a comprehensive approach to measurement assurance.
That means maintaining calibrated instrumentation, verifying calculations, reviewing meter factor trends, understanding operating conditions, and investigating discrepancies rather than simply accepting a passing report.
Because at the end of the day, the goal isn't just to prove a meter. The goal is to measure the product correctly.
How Upland Measurement Services Can Help
Upland Measurement Services provides liquid and gas measurement support to operators throughout Colorado, Oklahoma, Wyoming, Texas, and surrounding producing regions.
Our services include custody transfer meter proving, allocation meter proving, Coriolis meter verification, LACT maintenance, instrumentation calibration, measurement troubleshooting, and independent witnessing.
With operations based in Windsor, Colorado, and Mustang, Oklahoma, our team is positioned to support measurement programs across multiple producing basins.
Whether you need routine proving services or assistance investigating an unexplained measurement discrepancy, Upland Measurement Services is committed to helping operators maintain confidence in their measurement systems.
Upland Measurement Services
Proving is important. Measurement integrity is the objective.
