Proof Pressure Assessment
Engineering assessment of hydrostatic and pneumatic proof pressure, including structural load definition, elastic-plastic response, test configuration, acceptance and analysis support.
Purpose of a proof-pressure test
A proof-pressure test demonstrates that a pressure boundary can withstand a specified pressure above its normal service condition without unacceptable leakage, permanent deformation or damage. It is not intended to establish the ultimate burst pressure. The required proof level, duration, fluid, temperature and acceptance criteria come from the governing code, specification or programme and should not be replaced by a generic multiplier. Analysis supports the test by showing where the highest stresses and deformations are expected and by confirming that the proof condition itself does not introduce an unintended failure mode.
Hydrostatic versus pneumatic proof
Hydrostatic testing uses a nearly incompressible liquid and therefore stores relatively little elastic energy compared with compressed gas. Pneumatic proof can be necessary where liquid contamination, mass or drainage is unacceptable, but the stored energy and consequence of rupture are much greater. That distinction affects test safety planning even if the structural differential pressure is identical. For hydrotest, the weight of the liquid can add substantial dead load and support reactions, especially in large horizontal vessels. For either method, trapped gas pockets, elevation head and test temperature should be considered when defining the actual pressure field.
Structural model of the test condition
The test configuration may differ materially from service. Temporary blinds, caps, plugs, test manifolds, supports and transport cradles can create loads that do not exist in operation. A horizontal vessel may be completely liquid-filled during proof but gas-filled during service. Nozzles may be blanked, producing pressure thrust at closures. The analysis model should reproduce the proof arrangement rather than simply applying the proof pressure to the service model. Temporary equipment and support stiffness may need explicit representation where they influence local stress.
Elastic and permanent deformation
Many proof tests are intended to remain essentially elastic, while others permit tightly controlled local yielding depending on the design basis. The relevant acceptance criterion may include permanent set, dimensional change, leakage or visual indications. If nonlinear FEA predicts yielding, the analyst should assess unloading as well as loading to estimate residual deformation. Local elastic peaks at geometric discontinuities should be interpreted in the context of the specified proof criterion rather than compared automatically with a service allowable. The test requirement determines what constitutes acceptable response.
Pressure distribution and temperature
For liquid-filled vessels, pressure varies with elevation according to hydrostatic head. The pressure at the lowest point can therefore exceed the pressure measured at a top-mounted gauge. Material strength and toughness may also depend on test temperature. A cold hydrotest can change allowable stress and brittle-fracture considerations even though the pressure is temporary. Pressure instrumentation should be located and corrected so that the actual differential pressure at critical elevations is understood. Analysis should use the physically correct pressure distribution when the head is significant.
Instrumentation and observed response
Strain gauges, displacement measurements, pressure transducers and dimensional surveys can turn proof testing into useful validation evidence. Instrumentation should target locations where analysis predicts high membrane stress, local bending, flange rotation or deformation. Measurements should be compared with predicted elastic response before the proof peak is reached; unexpected nonlinearity, hysteresis or asymmetry can indicate slipping supports, contact change or local yielding. The objective is not to force test data to match the model but to use discrepancies to understand the real structure.
Acceptance and post-test inspection
A successful test is defined by the controlled test procedure. Typical evidence can include no pressure loss beyond permitted system effects, no leakage, no unacceptable permanent deformation, no crack-like indications and satisfactory inspection after depressurisation. Welds, nozzles, bolted joints and temporary closure interfaces often deserve focused examination. Where the pressure boundary is fatigue-critical, the proof cycle itself may need to be counted in the life assessment. The report should record actual pressure-time history, temperature, instrumentation and any anomalies.
Analysis as qualification evidence
Use analysis to establish that the planned proof condition is safe to conduct, to locate instrumentation, to predict expected response and to interpret any residual deformation. The model should be checked by hand calculations for hoop stress, end thrust and support loads and should reproduce the temporary test configuration. If the test pressure is close to a nonlinear limit, sensitivity to material properties and geometric tolerances is appropriate. The final assessment should clearly distinguish what was demonstrated by physical test from what remains justified by analysis.
Proof pressure is a specified qualification condition, not a universal percentage above operating pressure. Use the governing code or programme requirement.
Temporary closures and stored energy
Temporary test closures deserve the same structural attention as the permanent pressure boundary. Blind flanges, threaded plugs, bolted covers and test heads carry full pressure thrust and may have different load paths or materials from the service hardware. Their failure can be particularly hazardous during pneumatic testing because of stored energy. Verify bolt preload, thread engagement, flange bending and local attachment loads for the actual test arrangement. Exclusion zones, remote pressurisation and staged pressure holds are procedural controls rather than analysis results, but the engineering assessment should provide the load information needed to plan them safely.