A water draw procedure is the step-by-step field execution that establishes a prover's certified base volume by displacing water from the prover into certified test measures. Where the concept explains why the base volume matters, the procedure is the actual sequence a technician follows: fill and vent, let temperatures stabilize, launch the displacer, collect the water, read the measures, repeat, and correct. Every step exists to control a specific error, and skipping one shows up as runs that will not repeat. This page walks through that execution in order.
Water Draw Procedure in one line: A water draw procedure is the hands-on method for calibrating a prover: water is displaced through the prover's calibrated section as the sphere or piston travels between the detectors, collected in certified field-standard test measures, and read, with the run repeated several times. Temperature corrections bring the prover volume and the collected volume to common conditions, and the agreeing runs establish the certified base volume.
A water draw begins well before any water is collected, because most of the errors that spoil a draw are set up in the preparation. The prover and its connecting piping are filled completely with clean water and vented at every high point, since any trapped air or vapor is compressible and will make the displaced volume read wrong. The certified field standard test measures, the reference vessels the water is drawn into, are confirmed in date and clean, and the balance is checked if the draw is being done gravimetrically by weight rather than by volume. Isolation valves are proven not to pass, because a valve leaking around the calibrated section quietly changes how much water actually reflects the swept volume.
Temperature is the dominant variable in a water draw, so stabilization is a preparation step in its own right, not an afterthought. Water expands and contracts with temperature, the steel prover expands with temperature, and the two do not have to be at the same temperature as each other or as the collected water. The system is allowed to settle until the prover water, the prover wall, and the test measures are as close to thermal equilibrium as the setup allows, and thermometers are positioned to read each of them. Rushing this step is a classic cause of a draw that will not repeat, because a drifting temperature moves the answer between runs.
With the prover packed, vented, leak checked, and thermally settled, the flow path is arranged so the displacer will sweep the calibrated section cleanly and the displaced water will route to the test measures without leaking or splashing. Only when all of this is confirmed is the draw ready to run. The discipline of the setup is what lets the later runs be trusted; a draw taken on an unprepared prover can look repeatable and still be wrong.
A single run is one controlled displacement of the calibrated volume into the standards. The displacer is moved through the prover so it sweeps from the first detector to the second, and the water displaced during that travel is collected in the certified test measures. The technician reads the collected volume against the measure's scale, or reads the weight on the balance for a gravimetric draw, and records it together with the prover water temperature, the prover wall temperature, the temperature of the water in the measures, and the pressure. Each of those readings will be needed to correct the run, so none is optional.
One run is never enough, because a water draw is judged on repeatability, not on a single number. The run is repeated several times, and the collected volumes, once corrected, must agree with each other within a tight tolerance before the draw is accepted. Runs that agree closely say the whole chain, packing, venting, temperature control, reading technique, was under control. Runs that scatter say something was not: a lingering air pocket, a passing valve, a drifting temperature, or inconsistent reading of the measures. The technician does not average away a scattered set; they find and fix the cause and draw again.
Careful, consistent technique across the repeats is what makes them comparable. The measures are read the same way each time, to the same reference on the meniscus, drained the same way between runs so the same amount of clingage is left behind, and refilled cleanly. Temperatures are read at the same points at the same stage of each run. The aim is that the only thing changing between runs is ordinary measurement noise, so that when the runs do agree, the agreement is real evidence and not a coincidence of sloppy readings happening to land close.
The raw collected volumes are not the answer; they have to be corrected before they mean anything, because the water in the prover, the steel of the prover, and the water in the measures were all at their own temperatures and pressures. The corrections bring everything to a common reference: the thermal expansion of the water is accounted for, the thermal expansion of the steel prover is accounted for, the effect of pressure is accounted for, and the certified volume of the test measures at their own temperature is used rather than their nominal figure. Only after all runs are corrected onto that common basis are they compared for repeatability, and the agreeing set is averaged into the prover's certified base volume at reference conditions.
When a draw will not close, the correction stage often reveals where to look. Corrected runs that still scatter randomly point to a physical problem in the setup, air not fully vented, a valve passing, or unstable temperature, and the fix is back at the prover, not in the arithmetic. Corrected runs that trend steadily one direction point to a temperature that was still drifting through the session and never truly stabilized. A single outlier among otherwise tight runs points to a misread measure or a fumbled drain. The procedure is diagnostic: symptom in the numbers, cause at the prover, then draw again.
The certified base volume that a water draw produces is the number every future prove on that prover is referenced against, so it is a foundational record, not a routine reading. A cloud SCADA platform such as Merobix does not perform the physical draw, but it consumes what the draw produces: the base volume lives in the flow computer, and every automatic prove run afterward compares meter pulses against it. By trending prove results and meter factors across the fleet, the platform lets measurement staff see whether a freshly water drawn prover is producing consistent factors, flag when a prover's behavior suggests its calibration is due to be re established, and keep the whole chain from field standard to accounted volume visible from one place.
A water draw is judged on repeatability, not on a single reading. Repeating the run and requiring the corrected volumes to agree within a tight tolerance is how you prove the whole setup, packing, venting, temperature control, and reading technique, was actually under control. If the runs scatter, something is wrong, and averaging a scattered set would just hide the error rather than establish a trustworthy base volume.
Water and the steel prover both expand and contract with temperature, and the water in the prover, the prover wall, and the water collected in the test measures can all be at different temperatures. If temperatures are not stabilized and read carefully, the collected volume cannot be corrected to a common basis and the runs will not repeat. Allowing the system to reach thermal equilibrium before drawing is one of the most important steps in the whole procedure.
In a volumetric draw the displaced water is collected in certified test measures and read against their calibrated scale by volume. In a gravimetric draw the collected water is weighed on a calibrated balance and converted to volume using the water's density at its measured temperature. Gravimetric draws can achieve very high accuracy because weighing is precise, but both approaches follow the same fill, stabilize, draw, repeat, and correct sequence.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.