Pull-off Adhesion Test Principles

The pull-off adhesion test subjects a disc-shaped area of coating to a uniaxial tensile stress. A cylindrical metal dolly is bonded to the coating surface with a structural adhesive. Once the adhesive has cured, a portable adhesion tester grips the dolly and applies a tensile load perpendicular to the surface. The force increases until the weakest plane in the system fails, detaching the dolly together with whatever material remains bonded to it.

The pull-off stress is calculated by dividing the maximum force at failure by the nominal contact area of the dolly. For a circular dolly of diameter d, the contact area is πd²/4. This calculation assumes the load is distributed uniformly across the dolly face, which is an idealisation; in practice, stress distribution is influenced by dolly stiffness, adhesive uniformity, and substrate flatness.

The measured value represents the strength of the weakest plane at the specific location tested, under the specific loading conditions applied. It is not a material constant; it depends on the coating system, the substrate, the surface preparation, the test geometry, and the rate of loading. Two tests on the same surface can yield different values if the weakest plane varies between locations.

In operation, the tester is attached to the dolly via a coupling mechanism (typically a threaded connector or quick-release fitting). The instrument applies a steadily increasing tensile force, either through a manual hydraulic pump, a motorised actuator, or a hand-driven mechanism, depending on the tester design. The operator monitors the applied stress on a display or gauge until failure occurs.

The pull rate — how quickly stress increases — affects the result. Most standards specify a target pull rate (typically around 1 MPa/s for coatings on steel, or lower for concrete substrates). A faster rate can increase the apparent adhesion value for viscoelastic coatings; a slower rate can allow creep or stress redistribution, potentially lowering the measured value. Consistency of pull rate between tests is more important than the absolute value, provided it falls within the range specified by the applicable standard.

The typical output of a pull-off test consists of two elements: the numerical pull-off stress (in MPa) and the failure mode observed on the detached dolly face and the corresponding area on the surface. Both are required for a valid result. A pull-off value without a failure mode description is incomplete because the same numerical value can have entirely different implications depending on where the failure occurred.

Interpretation requires examining the fracture surface. If failure occurred within the adhesive (glue) used to bond the dolly, the test has not measured the coating system and the result is typically invalid. If failure occurred at the coating–substrate interface, the value represents the adhesion at that interface. If failure occurred within a coating layer (cohesive failure), the value represents the internal strength of that layer, not the interface adhesion.

• Material-dependent effects: Coating type, thickness, and internal structure influence where failure occurs and at what stress. Multi-layer systems can fail at any interface or within any layer. Coatings with high internal stress (e.g., thick single coats or coatings applied at low temperature) may show lower adhesion values even when the interface bond is sound, because the stored stress reduces the additional external stress needed to cause failure. Substrate porosity and surface profile affect the mechanical interlocking that contributes to adhesion on rough or porous substrates such as concrete or blast-cleaned steel.
• Environmental effects: Temperature at the time of testing affects both the coating’s mechanical properties and the adhesive’s bond strength. Cold conditions slow adhesive curing and can embrittle some coatings, while high temperatures can soften them. Moisture at the interface — from condensation, osmotic effects, or incomplete drying before coating — is a primary cause of adhesion loss in service and can produce low pull-off values at affected locations. Wind and dust during dolly bonding can contaminate the adhesive bond.
• Instrument/technique-specific effects: Tester alignment is critical. A pull that is not perpendicular to the surface introduces peel and shear components, reducing the measured tensile stress and potentially changing the failure mode. Self-aligning fixtures reduce but do not eliminate this effect on curved or uneven surfaces. The accuracy of the force measurement (pressure transducer, load cell, or mechanical gauge) directly affects the calculated stress. Dolly diameter must be known accurately because stress is inversely proportional to area; a 1 mm error in a 20 mm dolly diameter changes the calculated area by approximately 10%.
• Operator or setup issues: Inconsistent pull rate is the most common operator-induced variable. Jerky or accelerating pulls produce transient stress concentrations that can initiate failure prematurely. Inadequate coupling between the tester and the dolly (cross-threading, loose connection) can introduce bending. Testing on areas that are not representative — avoiding visible defects, testing only on the best-looking areas — produces results that do not reflect the overall coating condition.

Pull-off adhesion testing is applied in quality assurance during coating application (verifying each coat or the complete system), in acceptance testing before handover of coated assets, in failure investigation when coating detachment has occurred in service, and in research and development to compare coating formulations or surface preparation methods.

The pull-off value is often treated as an absolute measure of coating quality, yet this interpretation is misleading. A value of 5 MPa does not inherently mean “good” or “bad”; it must be compared against the specified acceptance criterion for the particular system and service environment. Equally problematic is the tendency to disregard glue failures. If the adhesive used to bond the dolly fails before the coating system, the test has measured the glue, not the coating. Reporting this as a valid adhesion result understates the actual coating performance.

It is also worth noting that higher values are not always better. In practice, extremely high pull-off values can indicate an unusually brittle coating or an adhesive that has bonded to a contaminated surface in a way that masks a weak underlying interface. Examining the failure mode provides the context needed to interpret the number correctly.

• Adhesion Test Failure Modes
• Surface Preparation and Dolly Bonding
• Verification & Checks for Adhesion Testers

Once the pull-off principle is clear, the next step is usually deciding whether you mainly need to choose a tester or tighten control over the whole test process.

• Select a Pull-off Adhesion Tester if the main question is which pull-off tester format fits the required range and workflow.
• Adhesion Test Setup and Validity if preparation, pull-rate control and result validity are the bigger concerns.