A resistance level printed on a band is a measurement claim, and a claim is only as good as the method behind it. Two suppliers can both print the word medium, both be honest, and still ship products whose force at the same stretch differs by a third. The remedy is not a nicer colour chart. It is a written definition, a calibrated rig and a tolerance rule that states exactly what happens when a batch falls outside it.
The four sections below are what a serious buyer should be able to point to in a purchase specification: the definition, the rig, the decision rule and the label. Where a studio group consolidates a mixed collection into one range, these numbers are fixed during custom development, before samples are signed off.
A level needs a method attached to the number
State four things together or none of them mean anything: the force in kilograms, the elongation at which it was measured, the width and thickness of the band, and the test method reference. Six kilograms at twice resting length in a forty-five millimetre width is a claim a laboratory can check. Medium is a word a lawyer cannot defend.
Textile and polymer test frameworks give both sides a shared vocabulary for elongation and force; the method library published by AATCC is one of the references buyers cite when they write that clause into an order.
Calibrate the rig before the first band is pulled
Half of all tension disputes are rig disputes. Fix these conditions in writing and repeat them on every batch:
- Gauge length of two hundred millimetres marked on the bench, with both ends clamped rather than hooked, so nothing slips.
- Crosshead speed held at five hundred millimetres per minute; slower pulls read low and faster pulls read high.
- Load cell checked against a reference mass at the start of each session, with the reading recorded.
- Room held at twenty degrees Celsius plus or minus two; latex softens noticeably in a warm room.
- Three conditioning pulls, discarded, before the recorded pull, so the band has taken its set.
Write the tolerance window and the decision rule
A workable window is plus or minus eight per cent of nominal force for a single unit, and plus or minus five per cent for the batch average. Sample the square root of the batch quantity plus one unit, minimum five, drawn from different cartons rather than the top layer. Accept the batch when every sampled unit sits inside the window. Hold and retest when one unit falls outside. Reject when two or more do, or when any single unit is off by more than fifteen per cent, because a single outlier that far out usually means a mixing error rather than natural scatter.
Match the level to the movement, then to the user
Build the level ladder from movements, not from colours. A long band used for standing rows and presses needs the lowest force in the range, because the anchor is far from the hand and the band is already stretched at the start. A short loop used for lateral walks and hip abduction needs the highest, because it is stretched very little. A ring sits between the two and is judged on compression rather than pull.
Then map users: rehabilitation and beginner classes take the lower third of the ladder, general studio classes the middle, and athletic conditioning the top. Write that mapping into the range sheet so a buyer ordering a refill knows which level replaces which.
Label wording that survives a complaint
Print six fields on the band or its hang card: nominal force and the elongation it refers to, the width, the batch code, the month of production, the care symbols, and a one-line instruction to inspect before each use. The batch code is what links a returned band to a test record.
- Say at twice resting length rather than at maximum stretch, which nobody can reproduce.
- Never print one number for a range of widths; change the width and the force changes with it.
- Add a short line that tension is measured, not guaranteed, and that a band showing nicks, tackiness or a pale stress mark should be replaced.
Labels are also where most after-sales arguments start, so keep the wording consistent with the specification sheet buyers receive; our production notes describe how the two documents are kept in step.
Control change between batches and reorders
Keep a golden sample of each level, sealed and dated, and retain three units from every batch. When a compound, a supplier of the latex sheeting or a width changes, treat it as a new product: retest, reprint the code and tell the buyer before shipment, not after.
A reorder that quietly drifts by a step is the most expensive failure in this category, because the complaint arrives months later from a class, not from a laboratory. Anyone setting up that control scheme for a new range can raise the tolerance question with our technical team before the first sample is moulded.
Frequently asked questions
Should a band level be given as a range?
Yes, when the same band is sold in several lengths. Print the nominal force for each length, since a longer band at the same width reads lower when pulled to the same absolute distance.
How long should test records be kept?
At least as long as the product is on the market, plus the warranty period. A complaint usually arrives with a batch code, and without the matching record the batch cannot be defended.
Does storage before shipment change the reading?
It can. Latex that has sat compressed in a warm warehouse reads low on first pull, which is why three conditioning pulls are discarded before the recorded one.
What if a buyer uses a different elongation?
Reprint the label for that market or supply a conversion table with the standard elongation stated. Comparing figures taken at different extensions is the most common source of false disputes.