How to Close Out a Well Test to Allocation
When the timed portion of a well test ends, the raw meter totals have to be turned into corrected rates and handed to the allocation model that splits the battery's sales volumes back to individual wells. This page is the closeout procedure. It walks the steps in order - read the totals, correct to reference conditions, compute the rates, sanity-check against history, and record the result - and explains why a sloppy closeout wastes an otherwise good test. It is for the operator or engineer finishing a routine production test.
Close out a well test to allocation in one line: To close out a well test to allocation, read the ending meter totals against the starting totals you recorded at the stable start, correct the oil, water, and gas volumes to reference conditions using the temperature, pressure, and gravity you logged, and divide by the test duration to get rates. Sanity-check the rates and water cut against the well's history, then record the result as the well's current test that the allocation model uses to apportion battery sales volumes.
Read the Ending Totals Against the Stable Start
The measured test is the difference between the ending meter totals and the starting totals you recorded when the separator stabilized, not the totals from when the well was first opened. Read the oil, water, and gas meter totals at the end of the timed window and subtract the start values. If you did not capture a clean stable-start reading, the test is compromised, because you cannot separate the steady flow from the stabilization flush.
Confirm the duration you divide by is the stable timed interval, matching the start you zeroed at in the setup covered by the guide on lining up a test separator for a well test. A common closeout error is dividing corrected volumes by the total time the well was flowing rather than the timed test window, which understates the rate. The window has to match the totals.
Correct Volumes to Reference Conditions
Raw meter volumes are at line conditions and mean little until corrected to a reference basis. Oil volume corrects for temperature and, depending on the meter, pressure and the observed gravity you logged; the gravity reading is the same measurement described in the guide on an observed API gravity reading. Water volume corrects similarly. Gas volume corrects to a standard pressure and temperature basis. Use the conditions you recorded during the stable test, not assumed defaults, because the correction is only as good as the inputs.
The water cut - the fraction of the liquid that is water - comes out of the corrected oil and water volumes and is one of the most scrutinized outputs of a test, because it drives how much of the battery's oil is allocated to this well. If the separator ran clean and the level held, this number is trustworthy; if there was carryover or blowby, it is not, which is why verifying level control during the test matters to the closeout.
Compute Rates and Sanity-Check Against History
Divide each corrected volume by the timed duration to get the oil, water, and gas rates, and compute the gas-oil ratio from the gas and oil rates. These rates are the test result. Before you accept them, compare them to the well's previous tests: a rate that has moved sharply with no operational reason, a water cut that jumped, or a gas-oil ratio that stepped are flags that either the well changed or the test is bad. A test that disagrees with history without explanation deserves a second look before it feeds allocation.
The reason this sanity check matters is that these rates become the well's allocation factor, the basis for splitting measured sales volumes back to the well, as described in the guide on what a well-test allocation factor is. A bad test that passes closeout unchecked will misallocate production for weeks until the next test, so the closeout is the last chance to catch it cheaply.
Record the Result and Feed the Model
Record the corrected rates, water cut, gas-oil ratio, test conditions, and duration as the well's current test, timestamped and tied to the well identifier. This record replaces the previous test in the allocation model and governs how the well is apportioned until the next test. An unrecorded or mislabeled test is a test that never happened as far as allocation is concerned.
A platform such as Merobix that has been trending the separator and meters through the test can carry the recorded conditions and totals straight into the test record, so the closeout reads from the same data that proved the test was stable rather than from hand-copied numbers. Keeping the test history in one place also makes the sanity check against previous tests a lookup rather than a paper search, and it ties the well-level tests to the battery reconciliation covered in the guide on production allocation in oil and gas.
Frequently Asked Questions
Do I divide test volumes by total flow time or the timed window?
By the stable timed window, not the total time the well was flowing. The stabilization flush before the timed start does not represent the well's steady rate, so including it understates the true rate. You read totals at the stable start and at the end of the window, take the difference, and divide by that window's duration. Matching the totals to the same interval you timed is essential for a correct rate.
Why correct well-test volumes to reference conditions?
Raw meter volumes are at line temperature and pressure and are not directly comparable or allocatable. Correcting oil and water for temperature, pressure, and observed gravity, and gas to a standard basis, puts every test on the same reference footing so rates can be compared across wells and tests and fed to the allocation model. The correction is only as good as the recorded conditions, which is why you log them during the stable test.
How does a well test become an allocation factor?
The corrected oil, water, and gas rates from the test become the well's current test result, and the allocation model uses those rates to apportion the battery's measured sales volumes back to each well in proportion to its tested contribution. The test result stands as the well's factor until the next test replaces it, so a bad test that passes closeout misallocates production until then, which is why the closeout sanity check against history matters.
Automation services
Need help turning this into a working system?
Merobix integrates SCADA, programs Allen-Bradley and Siemens PLCs, and designs and fabricates industrial control panels.
Meeting requests are reviewed before confirmation.