Quick answer. One question settles most of the brief: is there a thermoplastic fibre where the laser cuts and where the tape lands? Polyamide and polyester knits with elastane answer yes on both counts, which is why their edges seal and their bonds survive the care cycle. Cellulosic and protein fibres answer no — the edge chars, the loose fibres fray, and a hot-melt film finds no surface to wet. Section 2 grades each fibre on that basis.
1. The fibre decides, not the machine
A bonding line is set by three variables — temperature, pressure and dwell — and all three are tuned to the fabric in front of it. Change the fibre and the settings change with it. That is why two mills with the same tape and the same press can deliver one programme that holds and one that lifts, and why the fibre question comes before the machine question.
- Laser cutting is thermal. The beam removes material along the cut path. In a thermoplastic fibre the melt re-solidifies as a sealed bead a fraction of a millimetre wide, which is what removes the need for binding.
- Bonding is thermal too. A hot-melt adhesive film is activated between the layers and then sets as it cools. It adheres to a thermoplastic surface. Where there is no thermoplastic surface, there is nothing for it to hold.
- What that rules out. Cotton, cotton-rich blends, modal, viscose, lyocell and wool. A beam will cut them, but the edge chars rather than seals and the loose fibres fray — so there is no durable bonded edge.
- What it does not rule out. Blends where the thermoplastic component dominates. A 60/40 polyester–cotton can be cut and bonded; a 40/60 usually cannot, and the difference is not visible in a swatch.
2. The matrix, fibre by fibre
Read the third column as “how much of the bond survives the care cycle”, not as “does it stick on the bench”. The ratings assume a knit with elastane present and a declared bondable finish; a silicone softener moves any row down one grade.
| Fibre at the cut edge | Laser-cut edge | Hot-melt bond | Typical use | What to watch |
|---|---|---|---|---|
| Polyamide (nylon) 6 or 6.6 filament + elastane | Seals cleanly, soft hand | Strongest of the common fibres | Briefs, boxer briefs, thongs | Scorching at high power; the edge goes stiff and shiny |
| Polyester filament + elastane | Seals, slightly firmer edge | Good; needs a higher activation temperature than nylon | Sport, value programmes | Glazing — a hard edge that reads as cheap |
| Recycled nylon or rPET | Same as virgin | Same as virgin | Growing share of EU programmes | The content claim needs a transaction certificate, not a logo |
| Elastane as a major component (above roughly 20%) | Melts readily, can bead at the cut | Bonds, but the bond carries more of the stretch | High-stretch shaping styles | Heat damage to recovery; creep at the leg opening |
| Cotton | Chars, frays | None to poor | None — sewn, folded or covered edge only | A clean edge on cotton was covered or sewn, not sealed |
| Cotton-rich blend (cotton above roughly 50%) | Partial seal, fuzzy edge | Poor and inconsistent | None — sewn, folded or covered edge only | Batch-to-batch variation is the real risk |
| Modal, viscose, lyocell | Chars, frays | Poor | None — sewn, folded or covered edge only | Often proposed as “soft like nylon” — it is not nylon |
| Wool and wool blends | Chars | Poor | Rare in underwear | Not worth sampling for a bonded programme |
| Polypropylene | Melts | Poor without surface treatment | Rare | Low surface energy — a tape that holds on nylon will not hold here |
| PLA and bio-based polyester | Melts | Moderate, supplier-specific | Emerging | Ask for bond data on the actual lot, not the family datasheet |
| Nylon/elastane raschel lace | Laser or ultrasonic scallop | Good, on solid ground | Women’s briefs, thongs | The bond must land on a solid motif, not on open ground |
| Power mesh and tulle (nylon or polyester) | Seals, but the open area is fragile | Moderate | Panels and trims | Bond width must exceed the mesh cell, or the bond is a line of points |
3. Knit structure decides as much as fibre
Two fabrics can share a fibre content and a weight and still behave differently at the cut, because the loop structure sets how the edge behaves once the material is removed.
- Single jersey, circular. The default for bonded bodies. Predictable shrinkage, moderate edge curl, consistent bonding.
- Interlock and double knit. Heavier and denser, less curl at the cut, but it needs more pressure to wet the tape into the surface.
- Warp knit (tricot, raschel). Lace and moulded panels. The pattern sets where a bond can land, so the bond line is a design decision as much as a process one.
- Rib. High edge curl after cutting. The laser holds a clean line but the panel does not stay flat, and a curled edge shows under clothing — which is the one thing the construction exists to prevent.
The test costs nothing. Cut a 200 mm square, lay it flat at room conditions for 24 hours, and measure how far the corners lift. A curl you can see on a bench is a curl the wearer will see through a shirt.
4. What actually fails at the bond line
There are three failure modes, and only two of them are the bonding process.
- Adhesive failure. The tape releases from the fabric. Usually a surface problem: a finish, an oil, or a bond face that is not thermoplastic.
- Cohesive failure. The adhesive itself splits. Usually under-cure — temperature, dwell or cooling pressure was short.
- Substrate failure. The fabric tears before the bond does. This is not a bonding defect; it means the tape is stronger than the material, which is the outcome you want.
Peel strength is not one number. It is a curve, and the useful figure is the average over the steady middle section rather than the peak at the start. That is what ISO 11339:2022 measures on a flexible-to-flexible assembly, and it is the method to name in a purchase order when a supplier offers “strong bond” as a description.
5. The finishes that block a bond without anyone noticing
Most bond failures that appear after three washes are not process failures. They are surface failures, and they were decided at the dyehouse.
- Silicone micro-emulsion softener. The most common cause. It does exactly what it is bought for — lowers surface energy — which is also what stops an adhesive from wetting the surface.
- Fluorocarbon water repellent. Same mechanism, stronger effect.
- Heavy pigment print across the bond zone. The binder film sits between the tape and the fibre.
- Coning oil left on an unscoured greige. Invisible, and it migrates.
- Some anti-pilling and enzyme finishes. Less predictable — ask rather than assume.
The control is a written finish declaration on the fabric order — “bondable finish, no silicone softener on the bond face” — plus a three-piece trial before bulk. A supplier who cannot say what finish was applied cannot promise what the bond will do after ten washes.
6. Lace, mesh and foam: the hybrid cases
- Lace. Almost always nylon/elastane raschel, so it is bondable. What changes is geometry: the scallop is a pattern edge, and a bond that crosses open ground holds on threads rather than on fabric.
- Lace edges. Laser cutting and ultrasonic cutting both seal a scallop, but they produce different scallop shapes and a different hand. Specify the method and the repeat, not just the width.
- Moulded cups and bonded foam. The core is polyurethane foam bonded to a shell. It is a lamination, it is tested differently, and it should not be described with the same words as a bonded brief.
- Wording. If only the trim is bonded and the body is sewn, say so. A page that calls every smooth edge “bonded” gives a buyer nothing to specify against.
7. Test a candidate fabric before you commit
Five steps, in this order, on one fabric lot.
- Get the fibre content and the finish declaration in writing, including the elastane percentage by weight.
- Bond three pieces at the settings the mill intends to run in bulk — not at laboratory settings.
- Wash them to the care condition you will print, using the domestic washing and drying procedures of ISO 6330:2021. One cycle is not a test; run to the wash count you intend to claim.
- Run the T-peel test to ISO 11339:2022 on the washed, finished garment — not on a raw strip, and not on the tape datasheet.
- Check colour fastness on the same washed piece to ISO 105-C06. A bond that survives a shade shift is still a claim you cannot defend.
Compare the result against the figure in your own specification, not against a competitor’s. On the bonded programmes this site describes, the benchmark we work to is peel force on the finished garment after 30 washes.
8. What goes on the fabric specification
- Fibre content and blend ratio by weight, with the elastane percentage stated separately.
- Weight in g/m², with the tolerance you will accept on it.
- Knit structure: single jersey, interlock, warp knit or rib.
- Finish declaration, including “no silicone softener on the bond face”.
- Shrinkage after the care cycle, as a percentage rather than as “minimal”.
- Colour fastness values for the conditions you will print on the label.
- The bondable face, if the fabric is brushed or has a different face and back.
- Roll width and put-up, because a narrow roll changes the marker and the cost.
- The certification you require, by class: OEKO-TEX Standard 100 Class 2 is the class for products worn next to the skin.
- Whether the lot is a development lot or a bulk lot — bond behaviour is not always identical between the two.
If you are also ordering a sewn programme, the same discipline applies to the custom boxer briefs specification, where the equivalent of the bond line is the seam and the stitch density.
9. Where the matrix breaks
- “Laser-cut cotton”. If the edge is genuinely clean on a cotton garment, the laser is not what sealed it. Look for a covered edge, a folded and bonded tape, or a different cut entirely.
- Brushed inner faces. The brushed fibres are loose by design. They lift under a bond and they pill along the edge.
- Elastane below roughly 5%. The fabric barely recovers, so the bond carries the stretch instead of the knit, and it creeps.
- Space-dyed and two-tone yarns. Heat sensitivity varies between the yarns in the same fabric, so one side of the cut edge can take the heat differently from the other.
- Recycled content. Chemically the same as virgin, so the bond behaves the same. The claim is a separate question: ask for the transaction certificate and treat the fibre as a content issue rather than a performance one.
10. Bonded is not seamless — and the matrix does not transfer
The two constructions are routinely described with one word, and the fabric rules differ between them because the failure modes differ. A circular-knit seamless garment is knitted as a tube, so the fabric has to suit the machine programme; a bonded garment is cut from panels, so the fabric has to suit the laser and the tape. A fabric developed for one is not automatically a candidate for the other, and a quotation that treats them as interchangeable describes neither.
The practical consequence for sourcing is that a fabric approval does not travel. If you move a style from a seamless line to a bonded one — or the reverse — re-run the cut and peel checks on the new construction, on the same lot, before the specification is signed off.
11. How this is quoted, and what we will not claim
Unitex is a sourcing and trading company incorporated in Xiamen in 2013, working with partner factories that hold amfori BSCI and OEKO-TEX Standard 100 and that manufacture and bond the garments on their own lines. We are not the knitting or bonding line and we do not describe ourselves as a factory; the value we add is knowing which partner line should run a given fabric, and writing the specification that makes the answer checkable.
On the bonded line we quote nylon/elastane at 90/10 and 85/15, in the weights the mill can certify, and we ask for the peel figure on the finished garment before a bulk order is confirmed. If a fabric on your brief is one of the rows that does not bond, we would rather say so at quotation than at inspection. Our bonded and private-label women’s programmes are quoted the same way, because the fabric question is identical whether the garment is a boxer brief or a thong.
How do you tell a reliable supplier from a merely compliant one? By whether the answer to “what happens after 30 washes” arrives before the purchase order, with a test method and a figure attached. That is a question about the programme, not about the sample.
· Xiamen Unitex Trade Co., Ltd. — OEM & ODM underwear sourcing partner