The pull-off method is the most widely used quantitative technique for measuring the adhesion strength of coatings, adhesives, and bonded layers on metal and concrete substrates. By applying a controlled perpendicular tensile force through a bonded test dolly until detachment occurs, this test delivers precise adhesion values in MPa or psi in accordance with international standards such as ISO 4624, ASTM D4541, and GB/T 5210. This article explains the test principle, sample requirements, operating procedures, failure-mode analysis, and typical industrial applications of pull-off adhesion strength testing.
Technical Overview
The pull-off method, also known as tensile adhesion testing, is a quantitative destructive test used to determine the adhesion strength of a coating, membrane, or bonded layer to its substrate. A metal test dolly (stub) is bonded to the surface under test using a high-strength adhesive, typically a two-part epoxy. After the adhesive has fully cured, a portable adhesion tester grips the dolly and applies an increasing tensile load perpendicular to the surface until detachment occurs. The maximum tensile stress at failure, calculated as the breaking force divided by the dolly cross-sectional area, is reported as the pull-off adhesion strength in MPa or psi.
Unlike qualitative methods such as the cross-cut test (ISO 2409, ASTM D3359), which only rate adhesion on an ordinal scale, the pull-off method produces a numerical strength value that can be compared directly against product specifications, contract requirements, or qualification criteria. Equally important, the test identifies the exact plane and mode of failure, revealing whether the limiting factor lies at the substrate interface, within the coating itself, or within the test adhesive.
The most commonly applied standards for pull-off adhesion testing include:
- ISO 4624 — Paints and varnishes: pull-off test for adhesion
- ASTM D4541 — Pull-off strength of coatings using portable adhesion testers
- ASTM D7234 — Pull-off adhesion of coatings on concrete using portable testers
- ASTM C1583 — Pull-off strength of concrete substrates and overlays
- EN 1542 — Pull-off adhesion testing of concrete protection and repair products
- GB/T 5210 — Paints and varnishes: pull-off test for adhesion (Chinese national standard, technically aligned with ISO 4624)
Testing Samples
Anti-Corrosion and Protective Coatings on Steel
Epoxy, polyurethane, zinc-rich primer, and fusion-bonded epoxy (FBE) systems applied to steel structures, storage tanks, piping, and wind towers. These multi-coat systems are tested to verify both primer-to-substrate adhesion and intercoat adhesion between layers.
Coatings and Overlays on Concrete
Industrial floor coatings, epoxy and polyurethane screeds, cementitious repair mortars, crystalline waterproofing slurries, and protective coatings on concrete structures. Because concrete itself is comparatively weak, testing against standards such as ASTM D7234 and EN 1542 frequently reveals cohesive failure within the concrete substrate rather than at the coating interface.
Thermal Spray Metallic Coatings
Metallized zinc, aluminum, and zinc-aluminum coatings applied by arc or flame spraying to bridges, marine structures, and steel components. The pull-off method quantifies the bond strength between the sprayed deposit and the prepared steel surface.
Waterproofing Membranes and Sheet Systems
Bituminous, polymer-modified, and liquid-applied membranes on roofs, basements, and podium decks, where adhesion to the substrate governs long-term performance under wind uplift, hydrostatic pressure, and thermal cycling.
Specialty and Functional Coatings
Fire-resistive and intumescent coatings, tank linings, chemical-resistant coatings, and adhesive-bonded laminate assemblies in the aerospace, rail, and automotive sectors.
Testing Items
The core deliverables of a pull-off adhesion strength test are:
- Pull-off adhesion strength (MPa or psi) for each individual determination
- Failure mode and failure location for each determination, classified and reported per the standard
- Statistical summary of results: average value, minimum, maximum, and standard deviation
- Pass or fail judgment against the specified minimum adhesion requirement
- Optional documentation, including photographs of failed dollies and surfaces, test temperature, and dolly diameter
Testing Methods
Equipment and Test Setup
The test is performed with a portable adhesion tester, available in mechanical (screw-actuated), hydraulic, and pneumatic self-aligning designs, fitted with a calibrated load indication system. Standard dollies are made of aluminum or steel, most commonly 20 mm in diameter (cross-sectional area 3.14 cm²); other diameters such as 14 mm, 25 mm, and 50 mm are used to match the expected strength range and tester capacity.
Test Procedure
- Condition the test area, typically at 23 ± 2 °C and 50 ± 5 % relative humidity, in accordance with the governing standard.
- Lightly abrade the coating surface with fine abrasive paper and remove dust and contaminants with a suitable solvent so that the test adhesive can wet the surface.
- Abraid and degrease the dolly face, then bond the dolly to the prepared coating surface using a high-strength, low-creep two-part epoxy adhesive. Avoid adhesive squeeze-out at the dolly edge and cure for the full time recommended by the adhesive manufacturer.
- After cure, make a peripheral cut around the dolly down to the substrate using a hole saw or cutting template. This isolation cut ensures that only the coating directly beneath the dolly carries the load and prevents surrounding material from bridging the test area.
- Attach the tester actuator to the dolly, align the pulling axis perpendicular to the surface, and apply the tensile load at a controlled stress rate not exceeding approximately 1 MPa/s (150 psi/s) until failure occurs.
- Record the maximum force, calculate the pull-off strength as breaking force divided by dolly area, and inspect both the dolly face and the substrate to classify the failure mode.
Failure Mode Classification
Reporting the failure location is as important as reporting the numerical strength. The conventional notation identifies the following interfaces and layers:
- A/B — adhesive failure between substrate and first coat (often the governing adhesion value)
- B — cohesive failure within the first coat
- B/C — adhesive failure between first and subsequent coats (intercoat delamination)
- C — cohesive failure within a subsequent coat
- C/Y — adhesive failure between top coat and test adhesive
- Y — cohesive failure within the test adhesive
- Y/Z — failure between test adhesive and dolly (glue failure)
When failure occurs partly on multiple planes, the approximate percentage of each failure type on the dolly face is estimated and reported.
Result Validity and Reporting
- If the test adhesive fails before the coating system, the reported value is recorded as a lower-bound result (greater than or equal to the measured value) and the failure mode is noted.
- If the coating does not fail before the tester reaches its maximum capacity, the result is reported as exceeding that capacity limit.
- A minimum of three to five valid determinations per area or batch is customary, with results reported individually and statistically, together with dolly diameter, adhesive type, cure time, and ambient conditions.
- Test areas where dollies have detached are repaired by patch coating in accordance with the coating manufacturer's procedures.
Applications
Pull-off adhesion testing is routinely specified across industries wherever coating integrity is safety-critical or warranty-relevant:
- Oil and gas pipelines and storage tanks, verifying shop- and field-applied anti-corrosion coatings
- Marine and offshore structures, including ballast tanks and splash-zone coatings
- Bridges, steel buildings, and other infrastructure subject to aggressive service environments
- Concrete repair and structural strengthening projects, where adhesion of repair mortars and protective overlays must be demonstrated before acceptance
- Industrial and commercial flooring, confirming bond of epoxy and polyurethane systems to prepared concrete slabs
- Power generation and wind energy assets, including thermal spray coatings on towers and components
- Fire-resistive and intumescent coating installations requiring proof of substrate adhesion
- Coating qualification programs and dispute resolution in new construction and maintenance repainting
FAQ
What is the pull-off adhesion strength test?
It is a quantitative destructive test in which a dolly is glued to a coating, a perpendicular tensile force is applied through a portable tester, and the maximum stress at detachment is reported as the adhesion strength in MPa or psi, together with the observed failure mode.
Which standard should be used?
ISO 4624 and GB/T 5210 are generally applied to paints and varnishes on metal, ASTM D4541 to coatings on metal using portable testers, ASTM D7234 and ASTM C1583 to coatings and overlays on concrete, and EN 1542 to concrete protection and repair products. The contractual specification usually dictates the governing standard.
Why is the peripheral cut around the dolly necessary?
The isolation cut removes the contribution of the surrounding coating so that the measured stress corresponds only to the area directly under the dolly. Without the cut, results are typically biased high and non-comparable between operators and projects.
What does it mean if the glue fails instead of the coating?
A Y/Z glue failure indicates that the coating adhesion exceeds the strength of the test adhesive or the glue line. The coating adhesion is then reported as greater than or equal to the measured value; a stronger adhesive may be selected to force failure within the coating system.
Is the test destructive?
Yes. Each determination detaches a small circular area of coating, which must subsequently be repaired. The localized damage is minor relative to the structure, and test locations are chosen to be structurally and aesthetically acceptable.
How many test points are required?
Most specifications require a minimum of three to five valid determinations per representative area or production batch, distributed to cover edges, corners, main surfaces, and different application conditions, with all values and the statistical summary reported.
How are MPa and psi converted?
1 MPa is approximately equal to 145 psi. A common acceptance threshold for high-performance anti-corrosion systems is 5 MPa (about 725 psi) or higher, though project specifications vary and must always be checked.