Hydrostatic Pressure Testing of Piping: Methods & Criteria

Executive summary

A hydrostatic pressure test is how you confirm, before a pipe goes into service, that it holds pressure and has no leaks. You fill the section with water, bleed the air, raise the pressure above working pressure with a test pump, and hold it while inspecting joints and measuring make-up water. The two codes governing most installations are AWWA C600 (water mains) and ASME B31.3 (process piping). Each sets the test pressure, the hold time and the acceptance criterion. Here you will see how it is done step by step, which numbers apply, and why trapped air is the silent enemy of a clean test.

Who this is for

For project engineers, field supervisors, maintenance staff and commissioning leads who receive a new or repaired pipe and must certify it before pressurizing the system. If you ever wondered why we test at 1.5 times the working pressure, how long to hold, or what it means that a section "had allowable leakage but passed," this article puts those criteria in order with the code numbers.

The real problem in the field

Hydrostatic testing is often done "the way it has always been done": the pressure is raised until the gauge reads a round number, everyone waits a while, and the sheet is signed. Trouble shows up in two ways. First, by deficiency: the section is tested at too low a pressure or for too short a time, and a joint that would have failed in service passes the test without giving itself away. Second, through ignorance and excess: the air is not properly bled, so the section keeps a pocket of compressed air storing enormous energy. Unlike water, which is practically incompressible, compressed air turns a routine test into a real projectile hazard if something lets go. And there is a third, subtler problem: the testing allowance (the make-up water the code permits) is confused with an acceptable real leak, and good sections get rejected or bad ones approved by misreading the criterion.

Engineering fundamentals

A hydrostatic test rests on three quantities: the test pressure, the hold time and the allowable leakage. The test pressure is always higher than the working or design pressure, to give margin against transients and to force a defect to reveal itself.

For water mains, AWWA C600 sets the test pressure at no less than 1.25 times the working pressure at the highest point of the section and up to 1.5 times at the lowest point (the water column adds pressure downward). The leakage test is typically held for 2 hours, measuring the make-up water against the allowable leakage.

For process piping, ASME B31.3 (paragraph 345.4.2) sets the minimum hydrostatic test pressure as:

P_T = 1.5 · P · (S_T / S)

where P is the design pressure, S_T the allowable stress at test temperature and S the allowable stress at design temperature. When the test is run at ambient temperature and the design is also near ambient, the ratio S_T/S is essentially 1 and the formula reduces to the familiar rule: P_T = 1.5 · P. The code caps the S_T/S ratio at 6.5 so components are not over-stressed. The minimum hold for joint inspection is 10 minutes (paragraph 345.2.2).

The AWWA C600 allowable leakage is computed with the testing allowance:

L = N · D · √P / 7,400

with L in gallons per hour, N the number of joints in the section, D the nominal diameter in inches and P the average test pressure in psi. This allowance is not a tolerated leak: it accounts for legitimate effects such as residual air, water absorption by the cement lining and thermal expansion.

How to apply it step by step

• 1. Isolate the section: install blind flanges or test plugs at the ends and close the boundary valves. Never test against equipment that cannot take the test pressure.
• 2. Fill with clean water from the lowest point, leaving the vents at the highest point open so the air escapes.
• 3. Bleed all the air. This step is about safety, not cosmetics: an air pocket stores energy and ruins the leakage reading. Close the vents only when water comes out with no bubbles.
• 4. Pressurize with a test pump gradually up to the code test pressure (1.25-1.5× working for AWWA, 1.5× design for ASME B31.3).
• 5. Hold for the required time: minimum 10 min in ASME B31.3 for inspection; typically 2 h for the AWWA C600 leakage test.
• 6. Inspect and measure: check every joint and connection, and measure the make-up water needed to maintain pressure. Compare it against the allowable leakage.
• 7. Document: pressure, time, temperature, make-up water and result. An unrecorded test does not exist.

Worked example with numbers

Take a realistic case and compute the three quantities. Assumptions: ductile-iron water main, working pressure 150 psi (≈10.3 bar); a 1,000 ft (≈305 m) section of 8" (DN200) pipe with joints every 20 ft, i.e. 50 joints; average test pressure 200 psi (≈13.8 bar).

Step 1 — Test pressure (AWWA C600). At the highest point of the section, a minimum of 1.25 × 150 = 187.5 psi (≈12.9 bar). At the lowest point, up to 1.5 × 150 = 225 psi (≈15.5 bar). The average working pressure of the test lands on the order of 200 psi, the value we will use for leakage.

Step 2 — Allowable leakage (AWWA C600 testing allowance). Apply L = N · D · √P / 7,400:

• √200 = 14.142
• Numerator: 50 · 8 · 14.142 = 5,656.9
• L = 5,656.9 / 7,400 = 0.764 gph (gallons per hour), equivalent to ≈2.89 L/h.

Step 3 — Make-up water over 2 hours. Since the leakage test is held for 2 h, the maximum allowable make-up water is 0.764 × 2 = 1.53 gallons (≈5.8 L). If more than 1.53 gallons must be added over the 2 hours to maintain pressure, the section fails and the leak must be found.

Step 4 — Contrast with ASME B31.3. If the same service were process piping with a design pressure of 20 bar and a test at ambient temperature (S_T/S ≈ 1), the minimum test pressure would be P_T = 1.5 × 20 = 30 bar, held for at least 10 minutes while all joints are inspected. Here there is no "allowable leakage": the criterion is that there be no leak and no visible distortion.

The exercise makes the difference in philosophy clear: AWWA accepts a small make-up allowance because a buried main with many joints and a cement lining is never perfectly tight in the test; ASME B31.3, in welded process piping, demands full tightness for the hold time.

When it applies and when it doesn't

The hydrostatic test (with water) is the default and the safest option, because water is nearly incompressible and releases little energy if something gives. It applies to the vast majority of new or repaired water mains, process lines and fire-protection systems. It does not apply as-is when the pipe cannot bear the weight of the water (large diameters on light structures), when the service fluid reacts with water, or when the line cannot be dried afterward: in those cases a pneumatic test with air or inert gas is used, which ASME B31.3 permits under stricter conditions because of the stored energy. The practical rule: hydrostatic whenever you can; pneumatic only when hydrostatic is unfeasible and with reinforced safety controls.

Common mistakes

• Not bleeding the air. A pocket of compressed air stores energy, falsifies the leakage measurement and is the test's biggest safety hazard.
• Testing at working pressure instead of the code test pressure. With no margin above, a marginal defect never gives itself away.
• Confusing the testing allowance with permitted leakage. The allowance covers legitimate physical effects; it does not authorize a real localized leak.
• Holding for less than the required time or inspecting in a rush: slow leaks need time to appear.
• Testing against valves or equipment that cannot take the test pressure, damaging them.
• Not correcting for temperature: a change in water temperature during the test shifts the pressure and gets mistaken for a leak.
• Not documenting pressure, time, make-up and temperature. Without a record, the certification has no value.

Decision checklist

• Is the section isolated with blind flanges or plugs rated for the test pressure?
• Did you bleed all the air until water comes out with no bubbles?
• Does the test pressure match the code (1.25-1.5× working for AWWA C600; 1.5× design for ASME B31.3)?
• Did you hold the minimum time (10 min ASME; 2 h leakage for AWWA)?
• Did you compute the allowable leakage with L = N · D · √P / 7,400 before measuring make-up?
• Did you isolate valves and equipment that cannot take the test pressure?
• Did you record pressure, time, temperature, make-up water and result?

Frequently asked questions

What pressure is a pipe tested at?

It depends on the code. For water mains under AWWA C600, at least 1.25 times the working pressure at the highest point and up to 1.5 times at the lowest point. For process piping under ASME B31.3, the minimum hydrostatic test pressure is 1.5 times the design pressure (paragraph 345.4.2), with a temperature adjustment. You never test at the working pressure as-is: always above it, to gain margin.

How long is a hydrostatic test held?

ASME B31.3 requires a minimum of 10 minutes at test pressure to inspect joints and connections (paragraph 345.2.2). For water mains, AWWA C600 combines pressure with a leakage test that typically lasts 2 hours, measuring the make-up water against the allowable leakage. Actual time depends on the code, the diameter and field conditions.

How do you calculate allowable leakage in a hydrostatic test?

With the AWWA C600 testing allowance: L = N · D · √P / 7,400, where L is the allowable leakage in gallons per hour, N the number of joints, D the nominal diameter in inches and P the average test pressure in psi. That allowance covers residual air, cement-lining absorption and temperature variations, not a localized leak. If make-up exceeds it, the section fails.

SEMHYS tools

Before testing, you have to size the line properly. Our diameter calculator helps you pick the diameter by velocity and friction, which is the basis of the section you will later test. To know the real pressure the system must overcome at its operating point, the pump calculator computes the total dynamic head, useful for setting the starting working pressure. And if you need hydraulic calculation tools for your team, you can see them in the shop. Hydrostatic tests are always governed by the applicable code: use AWWA C600 or ASME B31.3 as the final reference.

References

1. AWWA (2017). ANSI/AWWA C600 — Installation of Ductile-Iron Mains and Their Appurtenances (test pressure 1.25-1.5× working; 2-hour leakage test; testing allowance L = N·D·√P/7,400).
2. ASME (2022). ASME B31.3 — Process Piping, paragraphs 345.2.2 (minimum 10-min hold) and 345.4.2 (minimum hydrostatic test pressure 1.5 × design pressure, with S_T/S ≤ 6.5).
3. AWWA (2018). ANSI/AWWA C605 — Underground Installation of PVC Pressure Pipe (allowable leakage formula for PVC pipe, with denominator 8,223).