Home › Field notes › ISO 12947 Martindale Abrasion for Bags: Differences, Choices and Limit

Modular Bag Pro · 2026-10-11 · 2,592 words · 12 min read

ISO 12947 Martindale abrasion rubs a mounted specimen against standard abradant under a specified pressure along a figure-of-eight path, producing a wear result that is comparable only when abradant type, pressure, mounting and stopping rule are all stated. The variables that carry the most weight are which abradant the request names, the pressure applied, whether a standard underlay is used, and how the end point will be judged, and our own programmes pair this evidence with release counting at AQL 2.5 against an order floor of 500 pieces. Choose Martindale where European apparel-derived specifications, the fabric supply chain or historical comparability favour it, and choose the rotary platform described in ASTM D3884 where a programme already holds rotary data. The limiting statement matters most: higher machine totals do not convert into months of service, and results describe flat rubbing on a coupon rather than structured contact on a finished body.

Martindale Under the ISO 12947 Parts: What the Motion Actually Does

The family published as ISO 12947 describes a specimen rubbed against a standard abrading surface while tracing a Lissajous figure. The intention behind that path is to attack material from many directions rather than one, so anisotropy in a weave should influence results less than it would under a single-direction stroke. That property is the family's principal claim, and it explains why apparel houses adopted it early.

The machine applies a defined pressure through a holder holding a disc of standard abradant, and the specimen sits over an underlay whose presence and construction are themselves specified. Cycles accumulate as the path repeats, and the operator inspects the surface at intervals determined by the stopping rule agreed in advance.

Because contact is continuous rather than localised to two wheel tracks, damage develops more evenly across the faced area. Coating systems therefore tend to show their failure by progressive dulling or cracking rather than by one aggressive line, which often makes the route kinder to coated constructions than a loaded wheel would be.

That difference should never be read as one route being stricter than the other. Each privileges certain constructions, and the same two materials can swap places depending on which mechanism dominates contact in real use. This is why selection between them is usually decided by who needs to read the data rather than by supposed rigour.

Verdict: Choose Martindale where many-direction contact, even damage development and European-document alignment serve the programme, and remember that the figure-of-eight path privileges coated and uniform constructions differently from a loaded wheel.

Abradant, Pressure and Underlay: Three Weightings That Change Everything

Ask any laboratory which setting causes the most disputes and the answer is almost always the abradant itself. The standard family describes more than one standard abradant, and each interacts differently with coating systems, filament fineness and surface finish. Naming the one used, with its own reference, is therefore non-negotiable on any request.

Pressure follows close behind. Higher pressure drives abradant into the surface and accelerates removal, but the relationship is not linear across constructions; a finish that performs acceptably at one loading can degrade abruptly at another. What matters for specification purposes is that the value appears on the report and never changes across a comparison.

Underlay is the quiet third term. Sitting beneath the specimen, it changes how load distributes, how heat moves away, and how much the material can deflect under the holder. Running one candidate with a standard felt underlay and another without it produces results that cannot be placed side by side, yet this is among the easiest mistakes to make when work is commissioned across two laboratories.

Mounting completes the list. Specimen slack, orientation relative to construction direction, and how the piece is seated all affect contact uniformity. Rather than leaving these to workshop habit, write them into the request, along with the number of specimens per condition.

Spec rule: Name the abradant type, pressure, underlay, mounting orientation, inspection interval, conditioning state and specimen count on every request, since a result quoting none of them cannot be repeated anywhere else.

Martindale Versus Rotary Contact: Which Mechanism Privileges Which Material

Between the figure-of-eight route and the loaded-wheel alternative, three practical differences decide outcomes: direction distribution, contact geometry, and what a programme can compare against historically.

Direction distribution favours the figure-eight route, provided the intention genuinely is to neutralise weave anisotropy. Rotary contact travels a circular track, so the same population of yarns is attacked repeatedly, and highly directional constructions can behave differently under it. Where grain effects genuinely worry a development team, running both routes occasionally reveals more than either alone.

Contact geometry is the more influential difference. Two loaded wheels concentrate pressure along narrow tracks, generating localised heat and cutting into coatings. A disc under distributed load spreads contact, which usually means coated and laminated constructions survive longer relative to uncoated counterparts than they would under wheels.

History is decisive in real programmes. A brand whose archive holds twenty seasons of one route's data gains more continuity by staying there than any theoretical argument could offer; switching routes mid-range destroys comparability for every earlier approval.

Takeaway: Stay on whichever route carries your historical record, and when switching is unavoidable, run both side by side on at least one lot so the archive can be bridged rather than abandoned.

Stating an End Point Across the Martindale Family

Parts of the ISO 12947 family describe specimen destruction, measurement of mass change and assessment of surface appearance. Selecting the appropriate part is therefore a scoping decision rather than a clerical one, since each answers a different question about the same machine.

Destruction endpoints — holes forming, or specified numbers of threads breaking — deliver unambiguous events. They suit applications where function is lost at a defined event, such as a lining expected to stay intact through repeated contact. Mass change suits constructions where material removal dominates, particularly heavily coated systems prone to gradual loss.

Appearance assessment suits consumer-facing surfaces, but it is only repeatable when the reference images or physical standards are retained and the judging rule states what counts as failure. Without them, one assessor's pass becomes another's reject, and the whole exercise collapses into opinion.

Inspection interval deserves particular attention. Damage does not announce itself politely; the difference between checking every few hundred cycles and every few thousand can shift a result substantially. Set intervals fine enough to resolve onset, then record condition at each one so that progress can be reviewed afterwards rather than guessed.

Coated, Laminated and Recycled Constructions: Where Rankings Invert

Finish matters more than base cloth in most bag applications, and this is where route choice becomes consequential. A polyurethane-coated face, a thermoplastic film laminate and an uncoated high-density weave can rank in three different orders between the figure-of-eight route and loaded wheels, because the two machines localise heat and stress differently. Programmes specifying heavily featured external carriers meet this first, since their contact zones alternate between coated faces and bare webbing edges.

Recycled content adds another variable worth naming explicitly. Post-consumer resin introduces variation in polymer chain length and occasionally in dye uptake, which can show up as spread between specimens rather than as lower average performance. Retesting frequency should reflect that reality, and traceability documentation under recognised chain-of-custody standards matters as much as the wear figure itself.

Laminate adhesion deserves separate consideration. A film bonded to a substrate may survive abrasion while gradually releasing from it, which no single wear route necessarily detects. Where experience suggests adhesion risk, pair wear work with adhesion checking and with inspection after conditioning.

Finally, remember colour. Dark colourways often run hotter under contact regardless of route, and some dye classes interact differently with coating and heat. Testing one colourway and claiming coverage for four is a shortcut that fails unpredictably, usually in the shade nobody tested.

Bottom line: Test the exact finished construction, including coating, laminate, colour and recycled content claim, and retest often enough to see the spread rather than only the average.

Extrapolation Boundaries: What Machine Cycles Cannot Tell You

The most damaging habit in this subject is publishing a high machine total as though it were a service-life assurance. Machine exposure is controlled, continuous contact against clean standard abradant at room conditions. Real use introduces grit, moisture, ultraviolet ageing, intermittent loading, folding at fixed creases and contact with hardware.

The second boundary concerns geometry. A coupon is flat, unsupported in one direction and free from nearby construction. Convertible configurations with panels that shift between roles concentrate wear where tension changes with position, which no flat specimen can reproduce.

The third concerns sequence. In service, ageing and wear interleave: ultraviolet embrittlement followed by abrasion produces different damage from abrasion followed by ageing. Running a clean sequence because it is easier to schedule produces optimistic conclusions, so where sequence matters, build it into the plan.

The fourth concerns combination with other properties. Material can rank well for wear while carrying a poor tear behaviour or losing hydrostatic performance after coating has been abraded, and the eventual complaint will concern whichever governed. Documenting interactions is better than optimising single figures.

Selection rule: State alongside every published cycle total that it is a controlled comparison of flat rubbing, and list the addendum: folding, edges, grit, ultraviolet sequence and hardware contact that the result does not cover.

Writing a Martindale Requirement Without Fabricating Numbers

Start from the failure the customer would complain about. If lining failure behind a strap would prompt a return, take that as the end point. If dulling on a visible face would, use appearance against a retained reference. Working from consequence produces an end point nobody has to invent.

Then specify the route fully: the standard family part, its cited revision, abradant type, pressure, underlay, mounting orientation, conditioning, inspection interval, stopping rule, specimen count and assessment references. Cost scales with specimen count and run length, so a fully specified request protects budget as well as accuracy.

Next, govern the standard. Decide who holds reference images or physical standards, how long they are kept, and what triggers re-qualification after a material, coating or supplier change. Change control is where abrasion requirements usually decay, because a reformulated finish quietly inherits an old specification.

Finally, agree consequences before they are needed: retest rules, who funds them, and what happens when results straddle the limit. Our production team schedules this evidence into the 6-10 working day sampling window precisely so that a late surprise does not force either a concession or a re-run.

Combining Evidence: Wear, Identification and Batch Discipline

Wear data never stands alone. It belongs beside identification work establishing that the right material arrived, water behaviour assessed on untreated cloth, sewn-assembly results for joins, and batch release counting at the end of the build. Removing any one of these layers leaves a plausible-looking gap.

Identification is the foundation. Grip strength reading tells you the lot is the lot that was qualified; without it, a wear result may belong to a different roll entirely. Wherever possible, order both from the same batch so they can be cross-referenced in one file.

Batch discipline then protects the decision through repetition. Incoming declarations supported by periodic retesting catch substitution or coating drift before cutting, and our own release gate stays at AQL 2.5 regardless of how strong the material file is, because wear evidence says nothing about workmanship variation.

Capacity planning shapes what is realistic inside one development cycle. Vetted partner facilities provide the 4,950 m² SGS-verified floor used for these programmes, holding 149 machines tended by 137 people deployed across 7 production lines, with monthly scheduling near 200,000 units. That capability, coordinated by QUANZHOU JUNYUAN BAGS, established 2014 under a founder who has worked the trade since 2004, still runs to one sequence per order: sampling, pre-production reference, production across 35-50 days.

Timing and Documentation: Keeping the Result Usable

Evidence arrives usefully only when it precedes a decision. Material comparisons belong in the sampling stage, occupying 6-10 working days for most builds and 12-15 where several constructions interact, so that findings can shape the drawing rather than annotate a fait accompli.

Once production begins, the 35-50 day window should already contain its evidence gates: lot checks before cutting, retained swatches for later comparison, then counting at AQL 2.5 against the order floor of 500 pieces. None of these gates is expensive individually; together they remove most pathways to a disputed shipment.

Documentation deserves a closing word. Report reference, specimen photographs, retained standards, request wording and decision notes should sit in one folder under the order number. Reconstructing that bundle from email months later rarely succeeds, and its absence is usually what turns a manageable question into a costly one.

Freight closes the calendar: 25-35 days by sea for planned volume, 5-8 days by air where recovery is required, and 3-5 days by courier for documents and swatch movements. Teams planning long-running programmes — particularly those refreshing travel platforms across annual seasons — benefit most from keeping one fixed evidence package rather than renegotiating scope each year.

Judgement: Keep one fixed evidence package — identification, wear, seams, joining methods, batch release — reported against the same order number every season, because comparability across years is worth more than any single improvement in test scope.