Testing standard
Adhesives — T-peel test for flexible-to-flexible bonded assemblies
Written and technically reviewed by Dak System Inc. engineeringLast reviewed
ISO 11339 bends the two unbonded ends of a flexible-to-flexible bonded assembly into a T, grips one arm in each jaw, and pulls. The result is the average peel force per unit width over a defined length of peel, not a maximum, because peeling is a continuous process rather than an event.
Two flexible adherends are bonded over a defined length with unbonded tabs left at one end. After curing and conditioning, the tabs are bent apart to form a T, one arm is gripped in each jaw, and the assembly is pulled at a constant rate so the bond separates progressively along its length. The force is averaged over a defined length of peel — excluding the higher force needed to start it — and divided by the bonded width to give a peel resistance in force per unit width. The trace itself is retained, because it carries information the average does not.
The running resistance of a bond to a crack travelling along it. This is a different question from lap shear, and the difference is the reason both tests exist. Shear spreads load over the whole bonded area; peel concentrates it at a moving crack front a fraction of a millimetre wide. A rigid, highly cross-linked adhesive can be excellent in shear and very poor in peel, because it cannot deform at the crack tip. Assuming one predicts the other is a common way to lose a bonded joint, and the standard notes, as ISO 4587 does, that the procedure does not provide design information.
Both adherends are flexible, so neither arm holds the peel angle. The angle is whatever the specimen adopts, which is why the geometry is named for its shape rather than an angle.
Peel and shear are not related properties. An adhesive can be excellent in lap shear and poor in peel, and a joint is often lost to the second while being designed against the first.
Average force over the defined peel length, divided by the bond width
Expressed as force per unit width, commonly N/mm. It is an average over a length, which is why the peeled distance is part of the report.
Excluded from the average
The force to initiate a peel is higher than to continue one and depends on how the tab was made, so including it reports the preparation.
Pneumatic vice action grips hold each arm of the T at a constant pressure through a long peel, which a screw grip does not — a peel test runs for hundreds of millimetres and a slipping grip corrupts the whole trace.
SpecificationsVery little force, measured properly. T-peel forces commonly sit between a few newtons and a few tens of newtons, so the load cell is chosen for that range rather than for the frame — an accuracy class applies only down to a stated fraction of capacity, and below it the class no longer describes the instrument. Long travel matters too: a peel runs for hundreds of millimetres, so the grips must hold without creeping for the whole of it, since a grip that slips corrupts the entire trace rather than one point on it.
Peeling too short a length, so the average is dominated by the start-up transient and describes how the tab was made rather than how the adhesive behaves. Including the initial peak in the average, which does the same thing more directly. Reducing the result to a mean and discarding the trace, which hides stick-slip — a saw-toothed record and a smooth one can average identically and mean entirely different things. And oversized load cells, which produce a plausible-looking trace with no traceable accuracy at these forces.
| ISO 11339 | ASTM D1876 | |
|---|---|---|
| Family | ISO | ASTM |
| Geometry | T-peel, both adherends flexible | T-peel, both adherends flexible |
| Result | Average force per unit width | Average peel load |
| Design information | Explicitly not provided | A comparative method |
Closely comparable methods for the same geometry. Keep to one within a data set, because specimen dimensions and the averaging length differ and both feed straight into the number.
It is the ISO T-peel test for bonded assemblies in which both adherends are flexible. The unbonded ends are bent apart to form a T, one arm is gripped in each jaw, and the assembly is pulled at a constant rate. The reported result is the average peel force per unit width over a defined length of peel. The current edition is ISO 11339:2022.
Because with two flexible adherends neither arm can hold a peel angle. The specimen adopts whatever angle the balance of forces produces, and that angle changes with the stiffness of the adherends and the toughness of the bond. The name describes the shape the specimen makes rather than a controlled geometry, and it is the reason T-peel results are only ever compared with other T-peel results.
Because peeling is a continuous process, not an event. The bond separates progressively along its length, so there is no single moment that characterises it — the force fluctuates as the crack front meets variations in the bond line. Averaging over a defined length describes the running resistance of the joint, which is the quantity of interest. A peak would describe one accident along the way.
Because starting a peel takes more force than continuing one, and how much more depends on how the unbonded tab was formed — where the release film ended, whether there is a bead of squeezed-out adhesive at the boundary. Including that peak reports the specimen preparation rather than the adhesive. Peeling far enough to establish a steady trace, and averaging only that portion, removes it.
Because two very different behaviours average to the same figure. A smooth trace means the crack front advanced steadily. A saw-toothed one means stick-slip — the bond arrested the crack, load built, and it jumped forward, repeatedly. The second is characteristic of a brittle bond line and behaves quite differently in service, and it is invisible once the trace has been reduced to a mean.
No, and the assumption that it does is how bonded joints are commonly lost. Shear loads the whole bonded area at once; peel concentrates all the load at a moving crack front a fraction of a millimetre wide. A rigid, highly cross-linked adhesive can be excellent in shear and very poor in peel, because it has no ability to deform at the crack tip. The two are tested separately for that reason.
A small one. T-peel forces on flexible adherends commonly sit between a few newtons and a few tens of newtons, and an accuracy class applies only down to a stated fraction of capacity. A cell sized for the frame rather than the force will produce a trace that looks plausible and carries no traceable accuracy at the level being measured.
Dak verifies against whichever standard the method names, and where a class applies our frames sit a class tighter than it asks.
| The method asks for | Dak supplies | |
|---|---|---|
| Capacity | Very low — T-peel forces are commonly a few newtons to a few tens of newtons | Load cells from 1 kg to 60 ton on the Series 7200, and 0.5 to 100 kN on the Series 9000 |
| Force accuracy | ISO 7500-1 Class 1 at the actual peel force, not at frame capacity | ISO 7500-1 Class 0.5 — a class tighter than the method asks |
| Gripping | Grips holding the two unbonded arms, one in each jaw, so the specimen forms a T | Wedge, vice-action, pneumatic and hydraulic grips, built to the specimen |
| Environment | 23 ± 2 °C standard laboratory atmosphere | 3009 series chambers, −150 °C to +400 °C — temperature only |
This page describes the method as practised. The governing text is the current edition from the issuing body. Tell us what you are testing and we will answer with the machine, the fixture and a quotation.
