November 3, 2026
A pressure test is the last chance to find out that something is wrong before the vessel leaves the shop — and the only test where the vessel is loaded in a way that mimics service. For an air receiver tank, the test is not a single generic step: the method (pneumatic or hydrostatic), the test pressure, the weld joint efficiency and the extent of radiography are all fixed at the design stage, and they are fixed together. Change one and the others no longer hold.
This note sets out how the two methods differ in code terms, why a pneumatic test forces a stricter NDT regime, and the six conditions we check before and during every hydrostatic test on the shop floor.
On a compliant drawing, the test method is an explicit design output: test medium, test pressure, hold time, and — by implication — the weld joint efficiency and NDT level the vessel must be built to. It cannot be renegotiated after the shell is closed, because the NDT that a pneumatic test requires has to be planned into the weld map, the RT film plan and the production sequence.
The choice between the two methods usually comes down to two things: pressure level and internal cleanliness.
For low and medium pressure vessels where the client has a high requirement on internal cleanliness, the design will normally specify a pneumatic test. The reason is practical rather than theoretical: a hydrostatic test leaves water in a vessel that is difficult to dry completely — in dome ends, in nozzle pockets, along the bottom of the shell. For a receiver feeding instrument air, electronics, food, pharmaceutical or medical processes, that residual moisture is a contamination source from day one, and on stainless steel it is also a chloride source sitting in a crevice. A pneumatic test leaves the internals dry and ready to passivate, purge or inspect.
The price of that cleanliness is risk. Compressed gas stores energy; water does not. A pneumatic test at the same pressure holds orders of magnitude more stored energy than a hydrostatic test, and a weld that fails under gas does not weep — it ruptures. The codes compensate for that risk in the only way available: by making the weld less likely to contain a defect in the first place.
Filling before a hydrostatic test. Water is introduced slowly from the lowest point and air is vented from the highest point — trapped air is the most common cause of a pressure drop that is not a leak.
The two methods carry different minimum multipliers, and the difference is deliberate:
In both cases the value is then corrected by the ratio of allowable stress at test temperature to allowable stress at design temperature, [σ]/[σ]t, whenever the two differ. For a vessel tested at ambient in a material whose allowable stress falls off with temperature, that correction can push the required test pressure above the nominal multiplier — which is why the number on the drawing is a calculated value and not a rule of thumb applied in the yard.
One caution that comes up repeatedly in export projects: the governing code decides the multiplier, and multipliers are not interchangeable. Under ASME BPVC Section VIII Division 1, the pneumatic test pressure is 1.10 × MAP per UG-100 and the hydrostatic test pressure is 1.30 × MAP per UG-99(b), both with the stress ratio applied. Under GB/T 150.4, the hydrostatic test is 1.25 × design pressure and the pneumatic test is 1.15 × design pressure. A vessel designed, stamped and documented to one code but tested to the other code's multiplier is a nonconformance, even if the test "passed". Read the drawing; do not carry a number from the last project.
| Item | Pneumatic test | Hydrostatic test |
|---|---|---|
| Test medium | Clean, dry air or nitrogen | Water (chloride controlled) |
| Minimum test pressure | 1.10 × design pressure (ASME UG-100) | 1.25 × design pressure (GB/T 150.4); 1.30 × MAP (ASME UG-99) |
| Weld joint efficiency | 1.0 mandatory | 0.85 minimum |
| Category A/B weld NDT | 100% RT, every film | 20% spot RT |
| Film acceptance level | Level II or better on all films | Level III or better |
| Stored energy at failure | Very high — rupture hazard | Low — leak/weep |
| Residual in vessel | None | Requires draining and drying |
Every receiver designed for a pneumatic test must be built with a weld joint efficiency of 1.0, which in practice means full radiography of all category A and B welds — longitudinal seams, circumferential seams, head-to-shell joints — with 100% RT and every film accepted at level II or better (GB/T 4730.2 or the equivalent acceptance criteria of the governing code). There is no spot-RT option on a pneumatic-tested vessel.
The logic is a straight trade of inspection for risk. With water as the medium, a defective weld reveals itself as a weep or a slow pressure loss while the shell is still intact. With gas, the same defect releases stored energy in milliseconds and the failure mode is a fragment throw, not a leak. Since the medium cannot be made safer, the weld population has to be made cleaner: no unrevealed defects, no partial inspection, no level III films accepted on the basis that "the hydro test will find it anyway".
A hydrostatic-tested vessel is allowed a joint efficiency as low as 0.85 with 20% spot RT on category A and B welds and level III acceptance — because the test medium itself provides the margin of safety. That is the standard trade-off, and it is legitimate. It is simply not available once the design selects a pneumatic test.
Weld examination at test pressure. Every accessible weld is inspected while the vessel is held at pressure — the inspection time is what sets the minimum hold time.
Pressure level and cleanliness decide the method; the hazard classification of the contained medium decides the NDT, and it can override both.
For these media, medium compatibility also constrains the test itself. A chlorine vessel should not be hydro-tested and left wet; residual moisture plus chlorine is an aggressive combination, and the drying step becomes as important as the test.
The following six conditions are checked on every hydrostatic test we run. Each one exists because of a specific failure mode seen in practice — none is administrative paperwork.
The test is not performed when the ambient temperature is below 5 °C. Water at low temperature combined with a cold shell raises the risk of brittle behaviour in thick-wall and low-alloy vessels, and it makes any leak hard to distinguish from condensation. Where the design specifies a minimum metal temperature, that value governs over the ambient figure.
All pressure gauges used for the test must be calibrated by an accredited laboratory and covered by a calibration certificate that is within its validity period — normally six months. The certificate number, gauge serial number and calibration date go into the test record. An out-of-date gauge invalidates the test regardless of the result, and this is one of the first documents a third-party inspector or notified body will ask for.
A calibration certificate for a test gauge, issued by an accredited metrology laboratory. Six-month validity; the certificate number is recorded against the test.
Hydrostatic and pneumatic tests both require at least two pressure gauges connected to the vessel, so that a single gauge failure cannot hide an overpressure or a pressure drop. The gauges should be of the same range and accuracy class, mounted so both are readable from the operator's position, and the readings recorded on both at each hold point. A test with one gauge is not a valid test, however good the gauge is.
Test manifold with duplicated gauges. Both readings are logged at the start and end of the hold period.
Testing in rain or in humid weather, or filling with water whose temperature differs significantly from the shell temperature, produces condensation on the outside of the vessel — and condensation running down a weld looks exactly like a weep. Since the whole purpose of holding at pressure is to find weeps, losing the ability to distinguish them is a real loss of test sensitivity, not a cosmetic issue. Where the schedule forces a test in poor weather, the vessel is shielded and the surface dried before inspection.
The hold period must be long enough to inspect every weld at full test pressure. In practice that means a minimum of 30 minutes, typically 30 to 45 minutes, set by the weld length and accessibility rather than by the clock. During the hold, the pressure must not fall: any drop is investigated before the test is accepted, and a drop attributable to trapped air or thermal stabilisation is documented rather than assumed. The vessel is inspected at test pressure, not after it is released.
The water used for the hydrostatic test must have a chloride content not exceeding 25 mg/L. For carbon steel this is good practice; for stainless steel it is absolute. Chloride is the initiator of chloride stress corrosion cracking in austenitic stainless steels, and a hydrostatic test is precisely the condition that promotes it — a wetted surface, residual chloride, and a period of standing water in a confined space. Deionised or reverse-osmosis water is used for stainless vessels, the chloride level is verified by a laboratory report, and the vessel is drained and dried immediately after the test rather than left wet overnight.
A water quality test report for the test medium. Chloride is the parameter that matters; the report becomes part of the vessel's documentation package.
The test itself takes under an hour. The evidence package is what determines whether it stands up:
Receivers staged in the test area. Vessels are tested one at a time, each with its own gauge set and its own recorded hold period.
Decide the test method on the drawing, not in the workshop. A pneumatic test buys internal cleanliness and pays for it with E = 1.0, 100% radiography and level II acceptance; a hydrostatic test accepts 0.85 and 20% spot RT in exchange for a benign failure mode. The medium being stored can raise the NDT requirement above either baseline, and for toxic duty a tightness test is a separate acceptance criterion. Then run the hydrostatic test under the six conditions above — temperature, calibrated gauges, two of them, weather, hold time, and water chemistry — and record all of it.
If you are specifying a receiver for a project where the test regime matters — instrument air for a clean process, hydrogen or natural gas duty, or a chlorine application — send us the design conditions, the governing code and the destination country. We will confirm the test method and NDT level that the code requires before the drawing is released, and quote accordingly. Our material and volume selection guide covers the sizing side; the ASME vs PED certification note covers which certificate the destination market will ask for.
