Abrasion resistance is the most quoted and least comparable number in bag material specifications. A supplier says a lining "passes 30,000 Martindale", another says a coated fabric "passed 500 Taber cycles", and a third cites 15,000 Wyzenbeek double rubs. All three statements can be true and still describe three completely different kinds of wear, measured on three different machines, under three different loads. Comparing them directly is meaningless.
For a buyer, the practical task is narrower than it first appears. Abrasion data is only useful when the machine, the abrasive medium, the load and the failure criterion are all stated, and when the threshold is tied to where the material actually sits on the bag. A trolley base panel and a zip puller sit in entirely different wear environments, and the same fabric can be correctly specified for one and wrongly specified for the other.
Why Abrasion Data Gets Misread
The confusion starts with the word itself. Abrasion is not a single physical event. Rubbing a surface against a woven wool cloth, against a rubber wheel loaded with grit, and against a reciprocating wire screen produce different failure mechanisms: fibre fibrillation, coating removal, and surface polishing respectively. Each machine is built to reproduce one of these, and each reports a number in its own unit.
The second source of confusion is the failure criterion. A Martindale result is not the cycle count at which the sample is destroyed. It is the count at which a defined change appears — the first visible thread break, the appearance of a hole, or a specified colour change on a grey scale. Two labs running the same sample can report different numbers simply because one stopped at "slight fuzzing" and the other at "two broken threads". Any specification that quotes a cycle count without naming the endpoint is not a specification.
What Each Machine Actually Measures
The three machines most often cited in bag specifications reproduce three different wear modes, and they are not interchangeable.
| Method | Motion | Typical standard | Reports | Best suited to |
|---|---|---|---|---|
| Martindale | Flat circular rubbing, Lissajous path, wool or wire abrasive | EN ISO 12947 / ASTM D4966; leather: EN ISO 17076 | Cycles to a defined change | Woven linings, canvas, suede, upholstery-weight fabrics |
| Taber | Rotating specimen under two loaded abrasive wheels | ASTM D4060; leather: ASTM D7255 | Cycles per mil of coating loss, or cycles to substrate | Coated fabrics, PU and PVC films, rigid panels, edge-painted leather |
| Wyzenbeek | Reciprocating, back-and-forth, wire screen or cotton duck | ASTM D4157 | Double rubs to thread break | Heavy wovens, webbing, canvas used on structure panels |
Reading across a row tells you what the number means. Reading down a column tells you what the test is for. A coated fabric quoted at 30,000 Martindale is not three times better than one quoted at 10,000 Taber — the two figures share no common scale.
Programme parameters at a glance
| Parameter | Commitment |
|---|---|
| Minimum order quantity | 500 pieces per style / colourway |
| Sampling lead time | 6–10 working days (12–15 days for complex constructions) |
| Bulk production lead time | 35–50 days after sample approval |
| Inspection standard | AQL 2.5 (ISO 2859-1), single sampling, general level II |
| Incoterms | FOB Xiamen — EXW, CIF and DDP available on request |
| Payment terms | T/T 30% deposit / 70% before shipment; L/C at sight for container loads |
| Verified production floor space | 4,950 m² (SGS-verified) |
| Team | 137 staff across 7 production lines |
| Equipment | 149 machines including computerised saddle-stitch and edge-painting units |
| Monthly capacity | 200,000 pieces |
| Entity | Founded 2014; principal has worked in bag development since 2004 |
Setting Thresholds by Where the Material Sits
The useful way to set an acceptance threshold is to start from the wear environment rather than from a competitor's specification. Split the bag into zones, decide what failure looks like in each zone, and only then pick a test and a number.
| Zone | Wear mechanism | Practical test | Indicative threshold |
|---|---|---|---|
| Trolley base and corner guards | Abrasion against tarmac, kerbs and car boots | Taber on the coated face; drop-and-drag rig | No substrate exposure after 1,000 cycles, H-18 wheel, 500 g |
| Body panel, soft bag | Rubbing against clothing and vehicle interiors | Martindale, wool abrasive | 25,000 cycles no thread break on wool; 15,000 on heavier canvas |
| Lining, main compartment | Repeated loading and unloading of hard objects | Martindale, 12 kPa head | 10,000–15,000 cycles, no hole, no colour transfer |
| Zip tape and edge binding | Sliding contact with slider and puller | Wyzenbeek or flex-fold plus tape abrasion | 15,000 double rubs without tape delamination |
| Shoulder strap webbing | Continuous rubbing against body and hardware | Wyzenbeek, cotton duck | 20,000 double rubs, edge fray under 2 mm |
These figures are starting points for negotiation, not universal law. What matters is that the threshold is written against a named zone, a named machine and a named endpoint, so that a failed lot can be argued about with evidence instead of opinion.
Load, Abrasive and Conditioning Change the Answer
Martindale heads are normally loaded at 9 kPa for apparel and 12 kPa for upholstery. Running the same fabric at the heavier load can halve the reported cycle count. Taber wheels come in grades — H-18 is a coarse calibrist wheel, H-22 is milder — and a specification that omits the grade is unenforceable. Wyzenbeek results depend on whether the abradant is wire screen or cotton duck, and the two differ by a factor that regularly exceeds three.
Conditioning matters as much as the machine. Textiles are normally conditioned at 20 °C and 65% relative humidity before testing; leather and coated fabrics behave differently when dry. A sample tested straight out of a shipping carton in a dry inland warehouse will often outperform the same material tested after a week at tropical humidity — and then underperform it in the field.
How Abrasion Tests Fail in Real Use
Abrasion resistance is a poor predictor of the failures that actually generate returns, and it is worth being explicit about the gap. The most common bag failures are seam failure, hardware pull-out, zip failure and coating delamination — none of which is an abrasion event.
- **Seam failure** — the thread or the needle holes give way before the material surface does. Test with seam strength (ASTM D1683) rather than abrasion.
- **Hardware pull-out** — the anchor tears through the panel. Test with a static pull rig on the actual attachment.
- **Coating delamination** — the PU film separates from the base fabric. Test with flex-fold and adhesion (peel) rather than surface abrasion.
- **Zip failure** — element wear and slider spring fatigue. Test with a cycle rig and element pull strength.
- **Edge paint cracking** — flex and impact, not rubbing. Test with repeated bend around a mandrel.
Specifying Abrasion in a Purchase Order
A defensible specification has five components, and leaving any one out hands the argument to the other side: the standard and its year, the machine configuration, the conditioning regime, the endpoint definition, and the sampling plan. "Martindale 25,000 cycles" fails all five tests. "EN ISO 12947-2:2016, 12 kPa, wool abradant, conditioned 24 h at 20 °C/65% RH, endpoint = first thread break, 3 specimens from the bulk roll, all three to pass" is a clause that can be enforced.
It is also worth fixing the consequence. Specify whether a failed abrasion result is a rejection, a re-test from a doubled sample, or a price adjustment. Buyers who leave this open find that a failed test becomes a conversation about the test method rather than about the material.
A wording trap to avoid
"Passes Martindale" without a cycle count usually means the material was tested, not that it passed a threshold. Ask for the laboratory report with the cycle count, the endpoint photograph and the abradant specification attached.
Working With Laboratory Reports
Third-party reports are worth what the traceability behind them is worth. Check that the report names the specific material reference — not just "PU coated fabric" — and that the reference matches the swatch you approved. Check the test date against the production date: a two-year-old report on a discontinued coating tells you nothing about the roll in your warehouse.
Where the number is borderline, ask for the raw data rather than the summary. A report showing three specimens at 24,800, 25,300 and 25,100 cycles is a material that sits right on the threshold and will fail intermittently in production. A report showing 31,000, 29,500 and 33,000 is a material with margin. The pass mark hides this distinction; the specimen table does not.
A Practical Screening Sequence
For a new programme, resist the urge to test everything. Screen with the cheapest relevant method first and only escalate where the zone justifies it.
- Confirm the material reference and the mill's own test data before paying for a laboratory.
- Run Taber on coated faces that will meet ground contact; run Martindale on linings and body panels.
- Add flex-fold and adhesion to any coated material that will be folded, rolled or stored compressed.
- Re-test after any change of coating supplier, base fabric or finishing line — a colour match is not a material match.
- Keep the approved swatch and its report together, and reference both in the tech pack.
Where Abrasion Testing Fits in a Programme
For a 500-piece programme, abrasion testing belongs at the sampling stage on the two or three zones that carry the most wear risk, not across the whole bill of materials. The cost of a handful of tests is trivial against the cost of a bulk lot that fails in the market, and the data gives the quality team a defensible basis for rejecting a substitution.
The programme team at QUANZHOU JUNYUAN BAGS routinely arranges abrasion, flex and adhesion testing alongside the standard AQL 2.5 inspection cycle, so that material performance and production quality are evidenced by the same paperwork trail.
Put a number on the wear zones
Send the material list and the zones you are most worried about. The programme team will propose which tests are worth running, and what threshold to write into the purchase order.
Frequently Asked Questions
What is a good Martindale score for a bag lining?
For a lining that carries hard objects in and out daily, 10,000 to 15,000 cycles at 12 kPa with no hole and no colour transfer is a reasonable working threshold. Lighter dress linings may pass at 8,000, but expect visible polishing within a season of heavy use.
Is Taber better than Martindale for coated fabrics?
They answer different questions. Taber applies a small, heavily loaded wheel to a rotating specimen and is good at showing how quickly a coating is worn through to the substrate. Martindale spreads the wear over a larger area and is better at showing surface change on a textile face. For coated materials used on ground-contact panels, Taber is usually the more relevant test.
How many cycles equals one year of use?
There is no reliable conversion. Cycle-to-time ratios depend on how often the bag is used and where it is placed. Treat abrasion results as a comparative screen between candidate materials rather than as a service-life prediction.
Do I need abrasion testing for a 500-piece order?
Not across the whole bill of materials. Test the two or three zones with the highest wear exposure, usually the trolley base, the lining and the strap webbing. That is enough to defend the material choice and to catch a substitution before bulk.
Why did my supplier's 30,000 Martindale fabric fail in the market?
Usually because the failure was not abrasion. Coating delamination, seam slippage and zip wear all present as "the bag fell apart" but are caused by different mechanisms. Ask which component failed and test that mechanism specifically.
What does "30,000 Martindale" actually mean if no endpoint is stated?
It means a test was run. Without an endpoint definition — first thread break, hole, or specified colour change — the number cannot be compared with any other figure, including the same supplier's figure from a previous year.
Can abrasion results be compared between laboratories?
Only when the standard, the machine configuration, the abradant, the load and the endpoint all match. Even then, inter-laboratory variation of ten to fifteen percent is normal. Treat a difference smaller than that as noise.
What conditioning do samples need before abrasion testing?
Textiles are normally conditioned for at least 24 hours at 20 °C and 65% relative humidity. Leather and coated fabrics should be tested in the condition they will meet in service, which for tropical markets means specifying humidity conditioning explicitly rather than accepting ambient lab conditions.
Does abrasion testing cover colour fastness?
No. Colour fastness is a separate family of tests covering rubbing, light, water and perspiration. Abrasion testing may note a colour change as its endpoint, but it does not measure staining or fading in the way a fastness test does.
Who should hold the approved swatch?
Both parties, with the same reference number written on the swatch, the laboratory report and the tech pack. If only the supplier holds it, a substitution becomes a matter of memory rather than of record.