Hydrostatic Test Pressure

Two pressures that get confused with each other: the most a flanged joint may be tested to, and the pressure a completed system must be tested at. Both worked through with tables.

By PipeData Editorial · Published 22 July 2026 · Updated 28 September 2026 · How the figures are checked

B16.5 test limit: 1.5 × the 100 °F rating
B31.3 hydrotest: 1.5 × design pressure, stress adjusted
Minimum hold time: 10 minutes, then examine
Rounding: Up to the next 25 psi increment

Short answer. Two different pressures get confused with each other. ASME B16.5 sets a limit: a flanged joint may be hydrostatically tested at up to 1.5 times the 100 °F rating of its class and material group. ASME B31.3 sets the test pressure for the completed system: 1.5 times the design pressure, adjusted upward when the test temperature is colder than the design temperature. The second must not exceed the first.

The B16.5 test limit

B16.5 allows flanged joints and flanged fittings to be put through a system hydrostatic test at up to 1.5 times the ambient rating for the class and material group, rounded up to the next 25 psi. The same figure is the shell test pressure for a flanged fitting, which is proof tested at the works. B16.5 does not require a flange on its own to be pressure tested.

Test limit = 1.5 × (rating at 100 °F), rounded up to the next 25 psi

ASME B16.5 pressure ratings at 100 °F, psig — the basis of the test limit.
Material group	Class 150	Class 300	Class 400	Class 600	Class 900	Class 1500	Class 2500
Group 1.1	285	740	985	1480	2220	3705	6170
Group 1.2	290	750	1000	1500	2250	3750	6250
Group 2.1	275	720	960	1440	2160	3600	6000
Group 2.2	275	720	960	1440	2160	3600	6000
Group 2.3	230	600	800	1200	1800	3000	5000
Group 1.9	290	750	1000	1500	2250	3750	6250
Group 1.10	290	750	1000	1500	2250	3750	6250
Group 1.15	290	750	1000	1500	2250	3750	6250

These are the ambient ratings for each material group. See the full P-T rating tables for the temperature curves.

Computed hydrostatic test limits, psig — 1.5 × the ambient rating, rounded up to 25 psi.
Material group	Class 150	Class 300	Class 400	Class 600	Class 900	Class 1500	Class 2500
Group 1.1	450	1125	1500	2225	3350	5575	9275
Group 1.2	450	1125	1500	2250	3375	5625	9375
Group 2.1	425	1100	1450	2175	3250	5400	9000
Group 2.2	425	1100	1450	2175	3250	5400	9000
Group 2.3	350	900	1200	1800	2700	4500	7500
Group 1.9	450	1125	1500	2250	3375	5625	9375
Group 1.10	450	1125	1500	2250	3375	5625	9375
Group 1.15	450	1125	1500	2250	3375	5625	9375

Computed from the ratings in the table above. Confirm against a current copy of ASME B16.5 before using these for acceptance.

The B31.3 system hydrotest

Once the system is built, ASME B31.3 paragraph 345.4.2 requires a hydrostatic leak test at not less than 1.5 times the design pressure. Where the design temperature is above the test temperature — which it almost always is — the test pressure is raised by the ratio of allowable stresses, because the material is stronger cold than it will be hot:

P_T = 1.5 × P × (S_T / S)
S_T = allowable stress at test temperature · S = allowable stress at design temperature

The ratio S_T/S is capped at 6.5, and the test pressure must not produce a stress above the material's yield point at test temperature — which is what actually limits it in practice on a hot line.

Worked example

A Class 300 carbon steel line, design pressure 500 psig, design temperature 650 °F, tested with water at ambient.

Base test pressure: 1.5 × 500 = 750 psig
Allowable stress for A106 Gr B: 20,000 psi at 100 °F and 17,300 psi at 650 °F, so S_T/S = 1.156
Corrected test pressure: 750 × 1.156 = 867 psig
Check against the flange: a Class 300 Group 1.1 flange is rated 740 psig at 100 °F, so it may be tested at up to 1.5 × 740 = 1110, rounded up to 1125 psig. The 867 psig test is within that, and the flanges do not limit it.

That last check is the one that gets missed. If the corrected test pressure exceeds 1.5 times the ambient rating of any flange, or the test limit of any valve or other component in the test envelope, the test cannot proceed at that pressure — either the component is removed and blanked, or the test is split into sections, or a lower test pressure is agreed with the owner under the code's provisions.

Practical requirements

Hydrostatic leak testing requirements, ASME B31.3.
Requirement	ASME B31.3 provision
Minimum hold time	At least 10 minutes at test pressure, and long enough after that to examine every joint for leaks.
Test fluid	Water unless there is a risk of damage from freezing or from adverse effects on the piping material.
Water chemistry	Chloride content is limited for austenitic stainless — commonly 50 ppm or less — to avoid stress corrosion cracking. Drain and dry promptly.
Test temperature	Kept above the material's ductile-brittle transition. Testing carbon steel with cold water on a cold day is a real brittle fracture risk.
Venting	The system must be vented at high points while filling. Trapped air stores energy and makes a failure violent.
Pneumatic testing	Permitted only where hydrostatic testing is impracticable, and requires a documented hazard assessment — stored energy in a gas test is orders of magnitude higher.

Paraphrased for orientation. The governing code text prevails — confirm against a current copy before planning a test.

Common questions

What is the hydrostatic test pressure for a Class 150 flange?

A Group 1.1 Class 150 flanged joint may be tested at up to 450 psig — 1.5 times the 285 psig ambient rating, rounded up to the next 25 psi. That is the most the flange allows. The pressure the system is actually tested at comes from the piping code and the design pressure.

How is the B31.3 hydrotest pressure calculated?

1.5 times the design pressure, multiplied by the ratio of allowable stress at test temperature to allowable stress at design temperature. The ratio is capped at 6.5 and the result must not yield the material.

How long must a hydrostatic test be held?

ASME B31.3 requires at least 10 minutes at the test pressure, and the test has to last long enough for every joint to be examined for leakage.

Why is chloride limited in stainless steel test water?

Because residual chlorides concentrate as the system dries and can initiate stress corrosion cracking in austenitic stainless. A limit around 50 ppm is common, with prompt draining and drying required.

Can I pneumatically test instead of hydrostatically?

Only where a hydrostatic test is impracticable — for example where the system cannot take the water weight or traces of water are unacceptable. A gas test stores vastly more energy and requires a documented hazard assessment.