Waterproof Bag Materials: TPU vs PVC vs PU-Coated Nylon

The Three Constructions, Defined by Where the Water Column Lives

The useful way to classify waterproof shell material is not by trade name but by where the barrier actually sits, and that is also the only honest basis for a waterproof fabric comparison, because that single fact predicts most of the behaviour a buyer will care about in year three. A TPU laminate carries the barrier as a solid film bonded to the back of the face fabric, so the film is the waterproofing and the fabric is the strength. A PVC tarpaulin carries it as a coating applied to both faces of a scrim, so the coating and the structure are the same body. A PU-coated nylon carries it as a coating applied to the inside face only, so the coating is thin, exposed to nothing, and entirely dependent on its own chemical stability.

That difference decides the ageing behaviour. A solid film has no plasticiser to migrate and no solvent to lose, so it ages by slow oxidation and it ages slowly. A coating carries the plasticiser that makes it flexible, and that plasticiser will leave the coating under heat and under hydrolysis, at which point the coating stiffens, then cracks, then lets water through a fabric that still looks intact. This is the reason a PU-coated nylon can hold 10,000 mm on the swatch card and 2,000 mm after two summers in a warehouse.

It also decides the seam. A homogeneous thermoplastic film or coating can be welded, meaning two surfaces fuse into one and the joint is as strong as the material. A coating on the inside of a woven fabric cannot be welded face to face, so the seam has to be sewn and then taped, and the tape is an adhesive joint sitting on a coating that is itself ageing. We publish the seam method alongside the shell on every quotation, because the shell number alone is not the product.

Water Column: What the Number Measures and What It Hides

Hydrostatic head is the number every buyer asks for first and it is the least informative of the numbers we measure. It reports the height of a water column a flat, unstressed, new sample of material resisted before three drops appeared on the far face. That is a real and reproducible measurement and it is useful for ranking materials of the same family against each other. It says nothing about what happens after ageing, nothing about a seam, nothing about a fold, and nothing about abrasion, all of which are the ways a bag actually fails.

The practical consequence is that two materials with the same printed figure are not equivalent, because the figure is measured on new material. We therefore never approve a shell on its as-received column. We approve it on the column retained after the hydrolysis sequence and after the cold bend, and we record both numbers on the specification. A material that holds 20,000 mm new and 9,000 mm aged is a 9,000 mm material for our purposes, and a buyer deserves to know that distinction before the order and not after the first returns season. The method detail is documented on our ISO 811 hydrostatic pressure page.

There is also a units trap in this category that is worth naming. Millimetres of water column, kilopascals and the older pounds-per-square-inch figures are all in circulation, and a supplier quoting one and a buyer reading another will disagree by a large margin without either being dishonest. We report in millimetres and we state the method alongside, and we ask every brand to put the method on its own specification sheet for the same reason.

The 70C at 95 Percent RH Hydrolysis Sequence

Hydrolysis is the reaction between water and the polymer that makes up the barrier, and it is the mechanism that ends the life of most coated waterproof fabrics. It is accelerated by heat and by humidity together, so a bag stored in a hot, damp container, a warehouse or a car boot is undergoing the same reaction that a laboratory chamber produces in days. We run the sequence on every shell we qualify, including on shells that a supplier has already supplied data for, because supplier data is usually new-condition data.

1. Cut the barrier in the production orientation, not as a loose swatch, and include at least one welded or taped seam in the panel
2. Condition the panel at 70 degrees C and 95 percent relative humidity in a controlled humidity chamber
3. Hold for 7 days, then remove and condition at standard atmosphere for 24 hours before any measurement
4. Visually inspect the barrier face and both sides of the seam for tackiness, bubbling, cracking or delamination
5. Flex the panel at a standard fold radius 20 times and inspect again at the fold
6. Re-measure the hydrostatic head to ISO 811 on the aged panel and record it against the as-received figure

The tackiness check at step four is the earliest and the most reliable indicator, and it is entirely a hand test. A coating that is hydrolysing becomes sticky and then develops a faint ammonia-like smell before it cracks, and both are detectable days before any change in the water column. We record it as a pass or fail on its own, separately from the retained column, because a shell that still holds water but has gone tacky will block on the roll and will pick up dirt on the shelf.

The table is illustrative of the pattern we see rather than a promise about any specific roll, and we run it on the actual lot before we quote, because the retention figure varies by more between suppliers than it does between material families. A well-made PU coating from a good mill outperforms a poorly made TPU laminate from a bad one. The point of the sequence is not to rank families by reputation, it is to catch the specific roll in front of us before it becomes five thousand bags.

Cold Bend at Minus 20 Degrees C: The Mandrel Test

Cold is the second ageing mechanism and it is tested differently, because the failure is mechanical rather than chemical. A barrier that is flexible at room temperature can be brittle below freezing, and a brittle barrier cracks at a fold the first time the bag is packed in the cold. That is a winter-sports, a motorcycle and a cold-chain failure, and it is invisible in a warm showroom or a warm sample room. We test it with a mandrel bend rather than by handling the material, because handling is subjective and a mandrel is not.

The procedure is simple and the interpretation is what matters. Condition the panel at minus 20 degrees C for four hours, bend it around a mandrel of a defined diameter within a few seconds of removal, and inspect the fold under magnification for cracking or for whitening that indicates a craze. Repeat at a smaller diameter to find the diameter at which the material first shows damage, and record that diameter. The diameter at first damage is the number we put on the specification, because it is comparable between materials and a pass or fail at one diameter is not.

• TPU laminates typically survive the smallest mandrel in the set with no visible change at the fold
• PVC tarpaulins stiffen measurably and may craze at the fold at moderate diameters unless a cold-flex package is specified
• PU-coated nylons are generally the best of the three at low temperature, because a thin coating cracks less than a thick film
• The base fabric matters as much as the barrier: a tight weave forces the barrier to bend through a tighter radius
• Any shell intended for a ski, motorcycle or cold-chain programme is qualified at minus 20 rather than at ambient

The last point is the one that decides the specification, and it is why we ask about the destination climate before we quote the shell. A programme shipping to a market with a real winter needs a cold-flex package in the barrier formulation, and that package has a cost and a lead time attached. A programme shipping only to tropical markets can spend that money on UV stability instead, which is the more relevant mechanism there. We make the split explicitly on the quotation rather than defaulting one way.

Weight, Hand and Packability per Square Metre

Weight is the specification that a retail customer feels first and the one a brand usually underestimates. It is reported per square metre of finished material and it should be reported with the coating or film weight broken out, because that is where the difference between two shells of the same denier actually sits. Two 420D nylons can differ by more than a hundred grams per square metre purely on the weight of the barrier applied to them, and the heavier one will feel stiffer, fold worse and weigh more in the finished bag.

Packability is the column that most often gets a product returned and it is the least often specified. A dry bag that will not roll to three turns is a dry bag that will not seal, regardless of what its shell can hold, and a bag that resists folding is a bag that a customer will compress badly and then blame for leaking. We therefore specify the shell partly from the closure: a roll-top needs a barrier that will roll tightly and hold the roll, and a zip lid does not. The two decisions are made together and never separately.

Abrasion and Puncture Are Two Different Numbers

Buyers ask for abrasion resistance and receive a single number, usually a cycle count from a Martindale or a Wyzenbeek machine, and then discover that a bag with a high cycle count has failed in the field. The reason is that abrasion and puncture are different mechanisms measured by different methods, and a material can be excellent at one and poor at the other. Abrasion is a rubbing failure on the face, and it is what a bag on a conveyor or against a deck experiences. Puncture is a separation failure between yarns, and it is what a bag carrying a tool or a fin experiences.

We measure coated-fabric abrasion to ASTM D751 and we add a separate puncture-propagation check, because the second is the one that turns a small nick into a hole. A ripstop grid is the standard mitigation for propagation and it works, but it works by limiting the tear once it has started and not by preventing the start. A heavier denier resists the start. A product specification that needs both uses a ripstop and a heavier denier, and that is a cost decision the brand makes with us rather than a technical one we make for them.

• Face abrasion, measured in cycles, decides how long the printed graphic and the face finish survive
• Puncture initiation, measured as a force, decides whether a sharp internal object starts a hole
• Tear propagation, measured as a force after initiation, decides whether a small hole becomes a big one
• Coating adhesion after abrasion decides whether the barrier survives the face being scuffed
• Abraded-then-tested water column is the only figure that predicts a real return, and almost nobody measures it

That last item is the one we would put on every specification sheet in the industry. A bag does not fail because its new material had a low column, it fails because its scuffed material has a damaged barrier. We abrade the face to a defined cycle count and then re-measure the column, and the retained figure after abrasion is the number we accept a shell on. This is described alongside the spray method on our AATCC 22 spray test page.

Weldability and How Each Material Takes a Seam

The seam is where a waterproof bag leaks, not the panel, and the seam method available to a designer is fixed by the shell. A thermoplastic barrier that is homogeneous through its thickness can be radio-frequency or hot-air welded, which fuses the two faces into one and produces a joint with no holes, no thread and no adhesive. A barrier applied to the inside face of a woven fabric cannot be welded, because the outer face has no thermoplastic to fuse, so the joint must be sewn, which perforates the material, and then sealed with a tape.

A welded joint and a sewn-and-taped joint behave differently in service and the difference is worth stating plainly. A welded joint is as waterproof as the material and it fails only if the weld parameters were wrong. A sewn-and-taped joint has two failure paths: the needle holes below the tape, and the adhesive bond of the tape itself, which ages on the same hydrolysis curve as the coating underneath it. This is why a sewn-and-taped bag that tested well when new declines faster than a welded bag of the same shell.

The practical rule we give a brand is this: if the product will be submerged, weld it, and if the product only needs to resist rain and spray, a sewn-and-taped seam on a good laminate is sufficient and cheaper. The mistake is specifying a submerged product on a non-weldable shell and hoping the tape holds, because a tape on a hydrolysing coating is a joint with two independent ageing mechanisms and neither is visible from the outside.

The Material Selection Matrix for Bag Shells

The matrix below is the tool we work from at the brief stage. It is deliberately organised by the environment the bag will live in rather than by the material, because the environment is what the brand knows and the material is what we know. Read down the left for the use case and across for the shell we would recommend, then treat the recommendation as a starting point for a discussion rather than a rule.

Two rows of that matrix are the ones brands argue with, and both arguments are reasonable. The hot-climate row excludes a PVC tarpaulin that is cheaper and welds beautifully, and a brand shipping to a temperate market can absolutely use it. The fashion row recommends the material with the shortest life, and we recommend it anyway because the product is not being bought for a five-year service life. What we insist on is that the trade-off is written down and acknowledged before the order, not discovered afterwards.

Cost Index and What Actually Moves It

We quote shell cost as an index against a baseline rather than as a figure in a published article, because a published price is wrong the moment it is printed. What does not change is the structure of the cost, and that structure is what a brand should understand before it asks for a cheaper material. The index below is relative, with the 600D TPU laminate baseline set at 100.

The second-order costs are where a programme loses money and they are all in the processing rather than in the material. A non-weldable shell adds a taping operation and a tape cost, and it adds an inspection step that a welded shell does not need. A stiff shell increases cutting waste on a nested marker. A shell that blocks on the roll in hot weather stops a line. We price the whole route rather than the roll, and we tell a brand when a cheaper shell will cost more by the time it is a finished bag.

Odour, Plasticiser Migration and Retail Complaints

Odour is the single most common cause of a negative retail review on a waterproof bag and it is almost never a defect. It is a property of the barrier chemistry, it is strongest on a fresh PVC tarpaulin, it is much lower on a TPU laminate, and it dissipates over weeks in open air. What turns a property into a complaint is packaging: a bag welded from a fresh PVC roll, folded, and sealed into a polybag within a day will arrive at a customer with the full odour load still inside it.

• Specify an open-air rest period between welding and packing, and put it on the production routing
• Vent the carton and the polybag, because a sealed bag keeps the odour in as effectively as it keeps water out
• Choose a TPU laminate for any product sold sealed in a retail box, because the complaint risk is materially lower
• Ask for an odour check at goods-in on any PVC programme, and reject a roll that is noticeably stronger than the approved sample
• Print a short care note telling the customer to air the bag, because it converts a complaint into an expected behaviour

Plasticiser migration is the same mechanism seen from the material's side and it has a second consequence beyond smell. A plasticiser that leaves a PVC coating does not simply disappear, it migrates to the surface, where it makes the face tacky, attracts dirt, and can soften or mark any printing applied on top. That is why a printed graphic on a PVC shell can fail by lifting while the shell itself is still perfectly waterproof. We test print adhesion after the heat sequence rather than before it for exactly this reason.

Chemical Compliance: REACH, RoHS and OEKO-TEX by Material

Compliance obligations sit on the brand and not on the mill, but the mill determines whether the obligation can be met, so the material decision is also a compliance decision. The three frameworks that matter for bag shells are REACH, which governs substances of very high concern in the European market; RoHS, which matters where a product incorporates electronic components such as a power bank or a heating element; and OEKO-TEX, which is a retail-driven textile label that many European buyers now ask for by default.

The practical difference between the three shells is in what has to be declared and evidenced. A PVC tarpaulin carries the heaviest declaration burden, because the plasticiser and any stabiliser package are both in scope and both vary by mill. A TPU laminate is the cleanest of the three to declare. A PU-coated nylon sits in between and is complicated by the solvent residues from the coating process, which are themselves in scope for some buyers. We ask for the declaration at the mill level and we hold it on file per lot rather than per programme.

For any product that will be sold to children, or that will be sold in the United States with a children's-product claim, the relevant framework is the CPSC regime and the CPC documentation that goes with it, and that requirement drives both the material and the hardware. We hold CPC documentation capability alongside CE, RoHS and SGS testing, and the site itself is audited to BSCI and certified to ISO 9001:2015. Materials are sourced and documented so that a brand can answer a retailer's questionnaire without a second round of testing.

Failure Modes Sorted by Material

Every shell has a characteristic failure, and recognising it is the fastest way to diagnose a returned product. The pattern below is what we see when a brand sends us a bag that failed in the field, and the diagnostic column is the question we ask first.

The diagnostic questions matter more than the table, because a failure is usually a specification error rather than a material defect. A TPU laminate that delaminated at a fold was specified with too tight a fold radius for its film weight. A PU-coated nylon that hydrolysed in eighteen months was specified for a climate it was never going to survive. Both are preventable at the drawing stage and neither is recoverable at the returns stage, which is why we insist on seeing the destination market before we quote the shell.

How We Qualify a Shell: The Approval Sequence

We do not accept a swatch card as evidence. Every shell that enters a programme goes through a fixed sequence, the sequence is run on the actual lot we intend to buy, and the results are held on file against the style number so that a reorder can be compared against the original qualification. That last point is the reason the sequence exists at all: a reorder that silently changes mill is the most common way a good product becomes a bad one.

1. Mill declaration and, where required, REACH, RoHS or OEKO-TEX evidence is collected for the specific lot
2. As-received hydrostatic head to ISO 811, with the method and the conditioning recorded
3. The 70 degrees C at 95 percent RH hydrolysis sequence, with the retained column recorded
4. The minus 20 degrees C mandrel bend, with the diameter at first damage recorded
5. Abrasion to ASTM D751, then the abraded sample is re-tested for water column
6. A weld or a tape trial on the production machine, with the joint strength recorded against the shell strength
7. A print adhesion trial where a graphic is specified, run after the heat stage rather than before it

The programme then runs as a normal build: a first sample in production material in 6-10 working days, or 12-15 where the construction is complex, and bulk production in 35-50 days from approval at 500 pcs per style per colour. Shipments are FOB Xiamen on T/T terms of 30 percent deposit and 70 percent before shipment, and every lot is inspected at AQL 2.5 under ISO 2859-1. The site behind the programme is SGS-verified at 4,950 square metres, seven production lines, 149 machines and 137 staff, with a monthly capacity of 200,000 pieces.

QUANZHOU JUNYUAN BAGS has run waterproof bag programmes since 2014, with the founding team in the trade since 2004, and we build through our validated production partner facility rather than brokering. If you are choosing between shells, send us the destination climate, the closure type and the contents the bag will carry, and we will return a matrix and a costed route rather than a generic recommendation. Briefs go to service@junyuanbags.com or by WhatsApp on +86 177 5002 0688, and the full range is on our products page.

Frequently Asked Questions

Q1: Which waterproof bag material lasts longest?

A: A TPU laminate lasts longest in most conditions because the barrier is a solid film with no plasticiser to migrate and no solvent to lose, so it ages by slow oxidation. In our 70 degrees C at 95 percent RH hydrolysis sequence a TPU laminate typically retains around 90 percent of its water column while a PU-coated nylon retains around 30 percent. The exception is a very cold climate, where a thin PU coating flexes better.

Q2: What is the difference between TPU laminate and PU-coated nylon?

A: In a TPU laminate the barrier is a solid film bonded to the fabric and it can be welded, so the seam needs no thread and no adhesive. In a PU-coated nylon the barrier is a coating applied to the inside face, so the seam has to be sewn and then taped, and the tape adhesive ages on the same hydrolysis curve as the coating beneath it. That seam difference matters more than the printed column.

Q3: Why does my waterproof bag stop being waterproof after two years?

A: In most cases the coating has hydrolysed, which is a reaction between water and the polymer accelerated by heat and humidity, and it is the normal end-of-life mechanism for a coated fabric. The coating stiffens, goes tacky, then cracks at a fold and lets water through fabric that still looks intact. Storing the bag hot and damp accelerates it, which is why a loft or a car boot is the worst place for one.

Q4: Is a higher hydrostatic head always better?

A: No, because the figure is measured on new, flat, unstressed material and it predicts nothing about the seam, the fold, the ageing behaviour or the abrasion resistance. A material holding 20,000 mm new and 9,000 mm after ageing is a 9,000 mm material in service. We approve shells on the retained column after hydrolysis, cold bend and abrasion rather than on the as-received figure printed on a swatch card.

Q5: Can PU-coated nylon bags be welded?

A: No, because the thermoplastic coating is applied to the inside face only and the outer face has nothing to fuse to. Those bags must be sewn, which perforates the material, and then sealed with a tape. A welded joint is as waterproof as the shell itself, while a sewn-and-taped joint has two failure paths, the needle holes under the tape and the adhesive bond of the tape, and the adhesive ages too.

Q6: Which material is best for a bag used below freezing?

A: A TPU laminate specified with a cold-flex package is the safest choice, and we qualify it at minus 20 degrees C on a mandrel rather than by handling it. A standard PVC tarpaulin stiffens and can craze at a fold below freezing unless a cold-flex package has been specified at the mill. A thin PU coating actually flexes well in the cold but is the weakest of the three on shelf life.

Q7: How do you test for coating hydrolysis?

A: We condition a production-orientation panel at 70 degrees C and 95 percent relative humidity for seven days, condition it back at standard atmosphere for twenty-four hours, then inspect for tackiness and cracking and re-measure the water column to ISO 811. The tackiness check is the earliest indicator and it is detectable days before any change in the column, which is why we record it separately.

Q8: Does a waterproof bag smell, and is that a defect?

A: Odour is a property of the barrier chemistry rather than a defect, it is strongest on a fresh PVC tarpaulin and much lower on a TPU laminate, and it dissipates in open air over weeks. It becomes a complaint when a bag is welded and sealed into a polybag within a day, which traps the full odour load. We specify an open-air rest period and a vented pack on PVC programmes.

Q9: Which shell gives the best abrasion resistance?

A: A 500D PVC tarpaulin gives the best face abrasion and the easiest wipe-clean face, which is why we use it for tool and site-work products. Abrasion and puncture are different mechanisms though, and a ripstop grid is what stops a small puncture from propagating into a slit. We measure abrasion to ASTM D751 and then re-test the abraded sample for water column.

Q10: What does the material mean for REACH and RoHS?

A: The mill determines whether the obligation can be met, so the material decision is a compliance decision. A PVC tarpaulin carries the heaviest declaration burden because the plasticiser and stabiliser packages are both in scope and both vary by mill, while a TPU laminate is the cleanest to declare. Where a product has electronic components, RoHS applies to those components as well as to the shell.

Q11: How much does the shell affect the unit cost?

A: Quoted as an index against a 600D TPU laminate at 100, a 420D PU-coated nylon typically sits at 55 to 70, a 500D PVC tarpaulin at 70 to 95 and an 840D TPU laminate at 135 to 165. The second-order costs often matter more, because a non-weldable shell adds a taping operation, a tape cost and an inspection step that a welded shell does not need.

Q12: Can you match a material I am already using?

A: Yes, and we prefer to, because matching an existing approved shell removes the qualification risk from a transfer. Send a physical sample or the mill reference and we will source the equivalent, run the hydrolysis and cold-bend sequence on it, and report the retained column against your current figure. Where an exact match is not available we will propose the nearest option and show you the difference.

Q13: What is your minimum order and lead time for a material change?

A: The minimum is 500 pcs per style per colour, and a first sample in production material takes 6-10 working days, or 12-15 where the construction is complex. Bulk production runs 35-50 days from approval. A new shell adds the qualification sequence before sampling, so we start that as soon as the mill reference is confirmed rather than after the tech pack is signed off.

Q14: How do you stop a reorder from silently changing material?

A: We hold the qualification results on file against the style number and we re-run the hydrolysis and cold-bend sequence on the reorder lot, then compare the retained column against the original figure. A reorder that changes mill without telling anyone is the most common way a good product becomes a bad one, so we treat the retained column as a controlled property and not as a one-off test.

People Also Ask

• Which is better for waterproof bags, TPU or PVC?
• What is hydrostatic head in waterproof fabric?
• Why does PU coating peel off?
• Is PVC tarpaulin safe for bags?
• How long does a waterproof coating last?
• Can you weld PU-coated nylon?
• What is the lightest waterproof bag material?