A reservoir liner should not be accepted for impoundment because it looks continuous from a distance. Seam integrity must be demonstrated section by section. A practical pre-fill inspection combines surface observation, non-destructive testing, review of welding records, and documented repair confirmation. Each method answers a different question: whether the seam is visibly sound, whether it is airtight or electrically continuous where applicable, whether it was made under controlled conditions, and whether every defect was closed properly.

The most reliable approach is to treat seam acceptance as a traceable process, not a final walkover. A reservoir may hold water for years, while access to a failed seam becomes difficult as soon as filling begins. Finding and repairing a weak fusion seam before impoundment is normally far less disruptive than draining water, exposing the liner, and rebuilding the affected area later.

Start with a complete seam map and inspection record

Before any field test begins, the inspection team needs to know exactly what has been welded. The seam layout should identify panel numbers, seam identification, weld direction, equipment used, operator, date, weather conditions, and the applicable test method. This is especially important around penetrations, sumps, corners, anchor trenches, pipe boots, transitions, and tie-ins to concrete structures. These locations often involve shorter seams, awkward access, changing geometry, or extrusion welds rather than long straight dual-track welds.

A seam map also prevents a common acceptance failure: testing representative lengths while leaving isolated seam sections unaccounted for. Every production seam should be either tested or included in a defined testing boundary that makes its coverage clear. If a seam was repaired, the repaired area needs its own identification and retest record. A repair is not complete merely because material has been placed over the suspected defect.

Review the installation records alongside the map. Field welding parameters should be consistent with approved trial seams for the liner resin, thickness, and site conditions. Temperature, speed, pressure, overlap preparation, and surface cleanliness all influence weld quality. A record showing stable machine settings does not prove seam integrity by itself, but it helps identify sections made during changing conditions, equipment adjustment, power interruption, or operator changeover.

Visual inspection comes first, but it is not the acceptance test

Walk every accessible reservoir liner seam before using pressure, vacuum, or electrical methods. Visual inspection is the fastest way to find defects that would otherwise make later testing inefficient or misleading. Inspect both sides where access permits, especially at slopes and transitions where wrinkles and stress concentration may be less obvious from one direction.

Look for fishmouths, folds trapped in the overlap, incomplete edge fusion, excessive squeeze-out, burns, holes, contamination, scratches, damaged edges, and areas where the liner was stretched or creased before welding. On dual-track fusion seams, the two weld tracks should appear continuous and reasonably uniform. A seam that changes width, has a visibly wandering weld path, or shows signs of overheating deserves closer review even if a later test does not immediately show a leak.

Do not confuse cosmetic appearance with seam quality. A neat seam can still have a weak bond caused by poor heat transfer or inadequate pressure. Conversely, an uneven-looking area may be acceptable only after it has passed the relevant test and matches the approved repair or welding procedure. Visual inspection is a screening tool and a way to direct testing effort, not a substitute for it.

Match the non-destructive test to the seam type

There is no single test suitable for every reservoir liner joint. The proper method depends on the liner material, seam configuration, access, and project specification. Long dual-track fusion seams are commonly checked by air-channel testing. Extrusion welds, patches, and single-track seams are more commonly examined using a vacuum box. Electrical leak-location methods can provide additional coverage in appropriate liner system configurations, but they do not replace seam-specific inspection where the specification requires direct seam testing.

Air-channel testing of dual-track seams

A dual-track fusion seam has two parallel welded tracks separated by an unwelded channel. This channel allows the seam to be pressurized from one end. If pressure behaves as expected throughout the designated hold period, the seam section is less likely to contain a continuous leak path. If pressure drops abnormally, the crew must locate the leak, isolate the affected length, and repair it.

Good technique matters. The channel must be clean and open, the test needles or fittings must not damage the surrounding weld, and both ends of the segment must be sealed in a controlled way. Testing excessively long sections can make small leaks harder to locate and may complicate interpretation when air temperature changes. Dividing the seam into manageable test lengths gives clearer results and more useful documentation.

Do not accept a seam simply because pressure was applied. The record should identify the seam section, the test start and finish conditions, the observed pressure behavior, the test operator, and the result. A sharp pressure loss is an obvious failure, but a gradual change should also be evaluated in context. A damaged fitting, poor end seal, changing temperature, or an actual seam leak can produce similar symptoms until the setup is checked.

Vacuum-box testing of repairs and extrusion welds

Vacuum testing is especially valuable for details that cannot be air-channel tested. The inspector wets the test area with a suitable solution, places a transparent vacuum box over the seam, and applies vacuum. Continuous bubbles emerging from a seam indicate a leak path. The transparent cover is important because it allows the inspector to distinguish a local bubble caused by trapped air at the edge of the box from bubbling that originates at the weld.

The method is sensitive to preparation. Dirt, damp debris, loose geotextile fibers, irregular wrinkles, or a poorly seated gasket can prevent the box from sealing and create unreliable results. Clean and dry the surface as required by the procedure, then overlap adjacent test placements so no portion of the seam is missed. Corners, pipe penetrations, and patches may require smaller boxes or more than one placement direction.

Vacuum testing should be deliberate around repairs. A patch has a perimeter seam, not just a center area. Inspect the full perimeter, including corners where extrusion material may be difficult to place evenly. When a repair covers a broad defect, examine the surrounding liner as well; installation damage sometimes extends beyond the visible tear or puncture.

Destructive samples confirm welding quality, but do not replace full seam testing

Trial welds and destructive peel or shear testing are used to demonstrate that the selected welding settings can produce an acceptable bond under current site conditions. Production samples, where required by the project quality plan, provide further evidence that those conditions remained under control. These samples should be taken from designated locations or from extension tabs so the reservoir liner itself is not unnecessarily compromised.

Destructive tests answer a different question from air-channel or vacuum tests. They assess bond behavior and failure mode in the sample; non-destructive tests check continuity along installed seam lengths. Passing one does not automatically compensate for failure in the other. For example, good trial samples cannot justify accepting a production seam that fails an air test, and a leak-tight seam may still require investigation if destructive samples show poor fusion behavior.

When field conditions change materially, make new trial seams before continuing. Wind, rain, liner surface temperature, dust, moisture, power quality, and a change in liner roll or thickness can alter welding behavior. This is a control decision, not paperwork. A reservoir installation often contains long production seams, so an uncontrolled change can affect more material than the crew realizes before the next inspection point.

Repair defects with a defined method, then retest the repair

Every failed area should be marked clearly, protected from being covered or forgotten, and entered into the defect log. The repair method depends on the defect. A localized hole may be patched with compatible liner material. A defective fusion seam may require removal and rewelding, or a cap strip may be used where the repair design permits. Extrusion welding is frequently used for patches and irregular details, provided the prepared surfaces are clean, dry, and properly abraded where required.

Before applying a patch, remove damaged or contaminated material until sound liner remains. Round patch corners to reduce the likelihood of edge lifting. The patch needs adequate extension beyond the damaged area, but excessive patch size can create unnecessary wrinkles and more perimeter seam to inspect. Once completed, the repair must receive the appropriate non-destructive test. Recording only the original failure and not the retest result leaves the acceptance file incomplete.

Equipment control supports this process. For projects welding compatible thermoplastic liners within the applicable thickness range, equipment with stable adjustment and visible operating parameters helps crews reproduce approved settings. The Geomembrane Welding Machine BGT-900 Series is designed for hot-wedge welding of materials including HDPE, LDPE, PVC, EVA, ECB, and PP, with adjustable temperature, speed, and pressure. Those controls are useful only when they are tied to trial-seam results, operator checks, and documented field testing; a welding machine cannot correct contaminated overlaps, wrinkles, or poor seam preparation.

Use a release checklist before water reaches the liner

The final inspection should occur when the liner remains fully visible and accessible. Do not schedule it after protective cover, ballast, concrete work, or reservoir hardware has restricted access to seams. The release decision is stronger when the quality file can show a complete chain from material identification to final repair acceptance.

• All liner panels, seams, repairs, and detail areas are identified on the as-built layout.
• Visual inspections are complete, including penetrations, anchor areas, corners, and interfaces with structures.
• Each seam has the required non-destructive test record, with clear pass, fail, repair, and retest status.
• Trial weld and destructive-test records correspond to the liner material and the work periods represented.
• Open defects, temporary patches, damaged protection layers, and untested sections have been closed before impoundment.
• Drainage, underliner, pipe, and appurtenant works have not punctured, abrased, or stressed the completed liner.

One overlooked point is coordination with other trades. A seam may pass inspection and then be damaged by pipe installation, anchor work, access traffic, or placement of protection materials. Establish a hold point after final seam acceptance and before the reservoir is released for filling. Any activity on the liner after that hold point should trigger a focused reinspection of the affected area.

What a defensible acceptance decision looks like

A defensible decision does not depend on a single passing test or a statement that the reservoir liner was “checked.” It shows that every installed seam was accounted for, the selected test method matched the seam design, failures were repaired under control, and repairs were retested. It also confirms that the liner was not damaged after inspection.

For quality and safety teams, this approach creates a usable record for handover and later maintenance. More importantly, it turns pre-impoundment inspection into a practical barrier against leakage: defects are found while they are visible, accessible, and repairable, rather than after stored water has made the consequences larger and the investigation more difficult.