Can 0.6 barg Detect Weld Leaks in a 100 m³ Tank?

A 100 m³ tank at 0.6 barg can expose a gross through-leak with a soap-film bubble test, potentially within minutes. It cannot, by pressure decay alone, prove that every weld is defect-free or reliably reveal a slow leak when temperature changes move the pressure more than the leak does. Treat leak detection, nondestructive examination, and structural proof testing as separate jobs.

Reject the usual quick fixes

Do not pressurize the tank, glance at one pressure gauge, wait an arbitrary time, and call the welds good. That approach fails because gas stores substantial energy, gauge resolution may be too coarse, and pressure follows gas temperature even when nothing leaks. Uneven heating from sunshine or cooling after compression can hide a leak or create a false pressure loss.

Quick fix	Why it fails	Use instead
Watch for pressure decay	Temperature and gauge uncertainty can exceed the leak signal.	Bubble-test accessible joints and temperature-correct recorded pressure.
Assume 0.6 barg proves weld integrity	A leak test finds leakage; it does not establish structural capacity.	Use an approved NDE and proof-test plan for the required acceptance objective.
Raise pressure toward 1.1 × MAWP	The correct test pressure depends on the governing design and test requirements.	Have the responsible engineer approve the pressure, medium, boundaries, and precautions.
Use penetrant examination on one surface	A relevant indication may open only to the opposite surface.	Examine both accessible sides of joint welds and bolted-joint areas.

Checkpoint: proceed only after the work order states whether the acceptance target is external leakage, surface-breaking weld defects, or structural integrity.

Set the pressure basis and test boundary

Resolve the units before connecting the compressor. The proposed pressure is explicitly 0.6 barg, while the 1.96 bar MAWP is not identified as gauge or absolute. Those values cannot be compared safely until both use the same reference. Likewise, 1.1 × 1.96 = 2.156 bar is only arithmetic; it is not authorization to apply that pressure, and the result inherits the unresolved pressure basis.

1. Read the nameplate, design records, repair records, and applicable test procedure.
2. Mark every component inside the pressure boundary: shell, nozzles, closures, temporary blanks, valves, fittings, gauges, and hoses.
3. Confirm that every temporary and permanent component is rated for the approved pressure and test medium.
4. Have the responsible engineer set the test pressure and acceptance criteria. Do not derive them solely from the MAWP multiplier mentioned in the job request.
5. Define an exclusion zone from a documented risk assessment. A separation described only as greater than 10 m from the workshop does not establish a safe distance.

Checkpoint: the approved procedure must show one unambiguous pressure reference, the complete boundary, the acceptance limit, and personnel controls.

Examine both sides before adding pressure

Perform visual examination first. Clean the weld and adjacent material enough to reveal cracks, porosity openings, undercut, mechanical damage, and leakage paths at attachments or bolted joints. Record the location of every indication so repairs can be traced and reinspected.

Dye penetrant examination is suited to surface-breaking discontinuities in compatible, nonporous materials. Apply the qualified process to both accessible sides of each joint weld; one-sided examination cannot characterize the opposite surface. Include relevant bolted-joint areas if they are part of the leakage boundary. Penetrant examination does not find every embedded flaw and does not replace a pressure test.

If the tank was previously in service or weld-repaired, stop treating the job as a simple fabrication leak check. Service history, material condition, repair method, and the governing inspection plan determine whether additional NDE is required.

Checkpoint: close or disposition every visual and penetrant indication before pressurization, and retain a weld map showing the inspected surfaces.

Prepare a controlled low-pressure leak test

Use hydrostatic testing for structural proof when the approved procedure permits filling the tank. Liquid stores far less elastic energy than compressed gas, so a boundary failure is generally less energetic. Filling, drainage, contamination, foundation loading, corrosion risk, and disposal still require review.

If pneumatic leak testing remains necessary, minimize exposed stored energy and personnel exposure. Fit a controlled pressure source, isolation valve, suitable relief protection, and pressure measurement with enough range and resolution to show the specified acceptance limit. Provide a gas-temperature measurement; ambient temperature alone may not represent the gas after compression.

1. Isolate the tank from equipment that is not part of the test.
2. Remove personnel from the exclusion zone and control access.
3. Increase pressure in controlled stages under the approved procedure.
4. Pause at each stage to check temporary connections and abnormal deformation, sound, or movement.
5. Isolate the compressor after reaching the approved test pressure, then allow the gas and vessel temperature to stabilize before starting a decay record.

Checkpoint: begin leak inspection only when pressure is stable, access is controlled, and the compressor can no longer mask leakage by feeding the tank.

Find leaks without trusting time alone

Apply an approved soap-film or bubble solution to accessible welds, joints, nozzles, and temporary closures. Persistent bubble growth identifies a through-leak; a gross leak may appear within minutes at 0.6 barg. Foam created during application is not by itself a leak indication, so watch for renewed or growing bubbles at a specific point.

There is no defensible universal hold time for a slow leak. The required duration depends on the acceptance leak rate, free gas volume, absolute pressure, gas temperature stability, instrument accuracy, and resolution. For a fixed-volume sealed tank, compare temperature-corrected readings using Pabs/T, with temperature in an absolute scale. A roughly constant Pabs/T indicates that observed pressure movement came mainly from temperature; a declining value warrants leak investigation, subject to instrument uncertainty.

Do not convert a pressure drop into a leak rate unless the calculation uses measured free volume, elapsed time, absolute pressure, absolute gas temperature, and a defined leak-rate reference condition. The nominal 100 m³ tank volume may not equal the free gas volume during the test.

Checkpoint: accept the leak test only against the written bubble-growth or quantified leak-rate criterion, not because the gauge looked steady for an arbitrary period.

Verify the repair from weld map to isolation point

1. Mark every confirmed leak and depressurize through the approved path before approaching or repairing it.
2. Repair under the applicable welding and inspection controls.
3. Repeat the required visual and penetrant examination on the repaired area.
4. Repeat the leak test at the approved pressure using the same instrument locations and acceptance criteria.
5. Inspect the entire test boundary, including temporary blanks and connections, so a leaking test fitting is not mistaken for a weld defect.
6. Record pressure basis, pressure, gas temperature, stabilization period, observation period, instruments, weld locations, indications, repairs, and final results.

Get production moving only after the approved acceptance criteria are met. Then replace temporary arrangements, restore the normal configuration, and independently verify valve positions, removed blinds, connected safeguards, and boundary tightness.

Checkpoint: the final record must connect every tested weld and boundary component to a passing result or an approved disposition.

FAQ