Home / Blog / Waterproof Technology

Seam Integrity Over Service Life

What seam integrity means after two seasons

A new seam is judged on three things: does it hold water under pressure, does it resist peeling when the panel is pulled away from the weld, and does it survive being folded. A seam at two years old is judged on the same three, but the answers have moved. The waterproof performance usually holds long after the peel strength has dropped, which is exactly why people are surprised when a bag that passed a dunk test last month opens at a corner this month.

The important distinction is between a leak and a separation. A leak is water finding a path through a defect that already exists: a pinhole, a fold, a grit particle in the weld. A separation is the weld letting go because the bond has weakened. Leaks are usually a packing or cleaning problem. Separations are a service life problem, and they are what this article is about.

In practice the two interact. A seam that has lost 30 percent of its peel strength is far more sensitive to abuse, because a corner that would have absorbed a snag now opens. This is why a bag in its fifth year should be treated differently from the same model in its first, even though both close the same way and both pass the same dunk test on a good day.

How a welded seam is built and what holds it together

Radio-frequency welding works because the molecules in a polar polymer such as TPU or PVC rotate and heat when they are placed in a high-frequency field. Two overlapping films under a shaped electrode heat at the interface, the polymer chains diffuse across the boundary, and the assembly cools under pressure as one continuous material. There is no adhesive and no thread, which is why a good weld is as strong as the film beside it.

The quality of that weld depends on four process variables: field energy, dwell time, electrode pressure and the condition of the film surface. Too little energy and the chains do not diffuse far enough, giving a weld that peels cleanly with a glossy face. Too much and the polymer degrades at the interface, giving a weld that looks perfect and fails brittle. Pressure controls the squeeze-out, and a contaminated surface blocks diffusion entirely.

What that means for service life is simple but often missed: the weld is strongest when it is made slightly below the degradation threshold with a long, cool cycle rather than at high energy with a short one. Production lines running at maximum throughput tend toward the second, and the difference shows up years later rather than on the day of the test. Full process detail is in our RF welding guide.

The fifth row is the one that surprises buyers. A taped seam looks reassuring and photographs well, but the adhesive is a separate material with its own ageing chemistry, and it is almost always the first thing to let go. A welded seam has no third material in the joint, which is the single biggest reason it outlasts tape by a factor of two.

Peel strength: the number behind service life

Peel strength is measured by cutting a 25 mm strip from the seam and pulling the two films apart at a fixed rate, usually 100 mm per minute, and recording the force. It is the best available proxy for seam integrity because it loads the joint in exactly the mode that fails in service: opening rather than shearing. A tensile test across the seam mostly measures the film, which is why it flatters a bad weld.

For a dry bag in regular use, 15 N per 25 mm is a working floor. Above it the seam absorbs ordinary abuse; below it, a corner caught on a rock or an over-tensioned strap can propagate a separation. That threshold is the reason a bag should be retired on seam condition rather than on appearance, and it is the number to ask for if you are writing a specification.

Test methods are published and should be named rather than implied. Peel and tensile methods for coated and laminated fabrics are maintained by ASTM International, and the equivalent international methods, including the accelerated ageing protocols used to generate a multi-year curve in a few weeks, are held by ISO. A claim without a method number is a claim you cannot check.

Three degradation paths: hydrolysis, ultraviolet and creep

Polyester-based TPU loses strength by hydrolysis: water molecules attack the ester linkages in the soft segment, the chains shorten, and the material becomes tacky and then weak. The rate roughly doubles for every 10 °C rise and it is fastest in warm, damp, enclosed conditions, which is exactly what the inside of a bag that was stored wet provides. Polyether-based TPU resists hydrolysis far better and is the right choice for a bag that lives in the tropics.

Ultraviolet attack works from the outside in and is mostly a surface problem, but the surface is where the weld margin is. A seam line is a slightly raised, slightly stressed band, and it takes UV at a slightly different rate from the flat panel beside it. The visible sign is chalking and loss of gloss along the weld; the hidden consequence is that the outer few micrometres of the joint embrittle and crack when the bag is folded.

Creep is the one people rarely attribute correctly. A strap anchor that has been under tension for a full season will have moved a measurable fraction of a millimetre, and the weld margin around it will have stretched with it. It does not look like damage, it looks like a bag that has been used. Releasing every strap in storage costs nothing and removes most of this path.

Heat and the temperature window that matters

Heat accelerates every degradation path at once, and the thresholds are lower than most people assume. Below 20 °C a stored welded seam loses very little strength in a year. Between 30 and 40 °C, which a bag in a closed vehicle or a locker on a tropical boat reaches routinely, the loss rate roughly doubles. Above 55 °C, which a dark bag on a dashboard reaches in an hour of sun, the soft segment begins to flow and a weld under any load will move permanently.

The load and the heat multiply rather than add. A seam held at 8 kg of strap tension and 45 °C will creep several times further than the same seam at either condition alone. This is the mechanism behind the classic failure where a bag comes out of a hot car with a visibly stretched weld margin at the strap anchor, and then separates at that anchor two trips later.

Cold is comparatively benign for seam integrity and mostly a handling problem. A TPU weld stays flexible well below freezing and does not become brittle until far colder than any paddler will experience; the risk is folding a stiff, cold film and putting a crease into it, which is a mechanical damage issue rather than a chemical one. Our temperature damage guide gives the full range by material.

Creep at the anchors and the loaded corners

Not all of a seam carries the same load. The margin around a strap anchor, a D-ring patch or a handle take the majority of the stress in a loaded bag, and they are where separations begin. A 25 mm strap anchor on a 0.5 mm film distributes load over perhaps 900 square millimetres of weld; the same strap on a 0.3 mm film without a reinforcement patch concentrates it into a fraction of that area and creeps accordingly.

The fix is a reinforcement patch welded under the anchor, which spreads the load into the panel rather than into the seam margin. In field returns, an anchor with a patch typically shows no measurable creep after three seasons, while the same geometry without one shows visible margin movement within one. It is one of the cheapest additions in the whole build and one of the easiest to specify.

Corners deserve separate attention because a corner is where two weld lines meet and where the peel mode and the shear mode combine. A well-designed corner is radiused rather than square, with the weld continuing past the corner rather than terminating at it. Terminating a weld at a corner creates a stress concentration and a short diffusion path, and it is the single most common origin point of a separation we see.

Field checks that find a failing seam early

1. Inflate the bag to firm pressure, close it, and brush soapy water along every weld line; bubbles mark a defect.
2. Run a fingernail lightly along the weld margin and feel for a step where the margin has stretched away from the panel.
3. Hold the weld up to a bright light and look for a whitened line, which is the film yielding before it separates.
4. Press a thumb along the inside of the seam and feel for a soft or tacky band, which indicates hydrolysis.
5. Flex the corner between two hands ten times and check the weld has not opened at the radius.

The soap test is the most useful and takes two minutes. Mix a few drops of dish soap in 250 ml of water, brush it along the seam with a small brush or a finger, and watch. A pinhole gives a slow stream of small bubbles; a separation gives a ridge of foam that grows. Mark anything you find with a pencil and re-test after the next trip before deciding whether it needs a repair.

Nails and lights catch what the soap test cannot, because they find a seam that is about to fail rather than one that already has. A whitened line under light is the film necking before it lets go, and a step you can feel with a nail is the margin creeping. Both are worth marking and watching; neither is yet a reason to retire the bag.

Repair windows: what can still be saved

Almost every seam failure is repairable if it is caught in the right window, and almost none are if it is caught late. A short separation on a flat panel, up to about 40 mm, can be re-welded on the right equipment and comes back near to original strength. A separation at a corner can be patched, but the patch is a get-you-home fix rather than a permanent one, because a corner carries combined peel and shear loading.

A seam that has gone tacky from hydrolysis cannot be re-welded at all. The polymer chains have shortened, the surface will not fuse, and no amount of heat or pressure will produce a bond. That bag is finished, and the useful action is to work out what caused the hydrolysis so the replacement is stored dry. Our seam repair guide covers the decision sequence in detail.

Keep the exception in mind: anything near the closure fold is different. The roll-top fold is the seal, and a repair across a fold line has to survive being rolled three times under pressure every single use. Nothing field-applied does that reliably, so a bag with a damaged fold should be retired even if the rest of the shell is sound.

Inspection intervals by use intensity

Set the interval from how hard the bag works, not from how old it is. A bag used six weekends a year in fresh water and stored dry needs one inspection a year, at the start of the season. A bag used weekly in salt needs one every three months. A rental or club bag doing 150 uses a season needs one every month, and a fleet operator should plan on retiring about 6 percent of the fleet per year on seam condition alone.

The inspection itself is the five checks above plus a written note. Recording what you found, even in a phone note, turns a vague impression into a trend, and the trend is what tells you when a bag is declining faster than its siblings. One bag failing early in a batch is a defect; several failing together is a specification problem worth raising with whoever built them.

Writing a seam life clause into a brief

• Name the seam type: RF welded, hot wedge, hot air or tape, and which for which joint.
• Name the peel strength required new, and the floor at which the bag is considered worn out.
• Name the test method and the ageing protocol used to generate any multi-year figure.
• Name the film chemistry: polyether TPU where hydrolysis risk is real.
• Name the reinforcement requirement at every strap anchor, ring and handle.
• Name the corner geometry: radiused, with the weld continued past the radius.
• Name the AQL level and the defect classes that are unacceptable at zero.

The last line is the one that gets a specification honoured. Sampling and acceptance are normally written against the standards held by ISO, and an AQL 2.5 plan with a zero-tolerance class for seal defects is the usual arrangement in this category. Our AQL inspection notes and the quality control overview set out how that is applied on a line.

Once the clause is written, validate it the same way you would validate hardware: pull samples from a pre-production run, weld-test them, and age a subset. Our own line has been produced since 2014 in an SGS-verified 4,950 m2 facility with 7 production lines and 149 machines, 200,000 pieces of monthly output, a 500-piece minimum, sampling in 6 to 10 working days and bulk in 35 to 50 days, FOB Xiamen.

What retires a bag: the seam decision rule

Use three questions in order. Is the film tacky, soft or chalky along the weld? If yes, retire, because no repair will hold. Is the separation on a flat panel and under 40 mm? If yes, repair and recheck. Is the damage at a corner, a fold line or an anchor? If yes, patch for the trip and replace the bag at the end of it.

Then apply the load rule as a cross-check. A bag carrying a sleeping bag and clothing is under almost no seam load in service and can safely be kept in use with a monitored repair. A bag used as a compression body, a raft lash point or a canoe thwart bag carries real seam load every trip, and any separation on it is a retirement, not a repair. Our retirement guide covers the wider decision.

Finally, retire on trend rather than on a single reading. Two consecutive inspections showing the same defect growing means the joint is actively failing, and the question stops being whether it will open and becomes when. A bag that opens at a seam on the water costs the contents, which is always worth more than the bag.

How a seam claim should be tested

A credible seam claim has four parts: the seam type, the new peel value, the ageing protocol used to produce any multi-year figure, and the sample size. Most published claims have one or none. If a listing says a seam is welded and lifetime guaranteed, ask which of the four is missing, and you will usually find it is all of them.

Accelerated ageing is legitimate but it must be described. Heat ageing a sample at 70 °C for seven days is a recognised screen for thermal stability, and it does not predict five years of coastal use; it ranks materials. A claim that says equivalent to five years without naming the protocol, the temperature and the acceptance criterion is a marketing statement, and the way to handle it is to ask for the raw numbers.

If you are comparing two constructions, the cheapest reliable comparison is your own. Buy one of each, use them identically for a season, then cut a 25 mm strip from an inconspicuous seam on each and pull them apart by hand against a luggage scale. It is crude, it destroys two bags, and it tells you more than any listing. Then compare what you learned against the full product line and against our seam failure notes before the next order.