Pull-Out Strength Testing Methods: Standards, Procedures, and Applications

Pull-out strength testing is the primary method for verifying the tensile anchorage capacity of anchor bolts, reinforcing bars, and mechanical fasteners embedded in concrete, masonry, and other substrates. This technical guide explains the governing international standards, specimen types, measured parameters, step-by-step test procedures, and typical applications of pull-out strength detection methods. Whether the goal is design verification, product qualification, or in-situ quality control, accredited third-party laboratories rely on these standardized approaches to evaluate pull-out resistance with traceable accuracy.

Technical Overview

Pull-out strength, also referred to as withdrawal strength or extraction resistance, is the maximum axial tensile load that an anchored component can withstand before it is extracted from its base material or before the load-transfer interface fails. This parameter is decisive for post-installed anchors, cast-in anchor bolts, reinforcing bars, mechanical fasteners, threaded inserts, and similar load-bearing connections. Pull-out testing quantifies the ultimate capacity, the load-displacement behavior, and the governing failure mode, providing the data required for design verification, product qualification, and conformity assessment.

Physical pull-out testing is essential because computational anchorage design alone cannot fully capture real-world influences such as substrate strength, drilling quality, hole cleaning, installation torque, curing condition of adhesives, and edge effects. The most widely applied reference standards include ASTM E488 for anchors in concrete and masonry, ASTM C900 for pullout testing of hardened concrete, ASTM D1761 for fasteners in wood, ASTM F606 for mechanical testing of threaded fasteners, EN 1992-4 for the design of fastenings in concrete, GB 50367 for strengthening of concrete structures, and ISO 6892-1 or GB/T 228.1 for tensile characterization of the metallic components themselves.

Testing Samples

Post-Installed and Cast-In Anchors

Specimens include wedge anchors, sleeve anchors, undercut anchors, heavy-duty torque-controlled anchors, and chemical (adhesive or capsule) anchors. The base materials are typically normal-weight or lightweight concrete slabs and blocks, masonry units, or natural stone. Specimen geometry, effective embedment depth, edge distance, and spacing are defined according to the relevant product standard or the project specification.

Reinforcing Bars and Strengthening Systems

Post-installed reinforcing bars used for structural strengthening, bonded-in rebar connections, and lap-splice specimens are tested to evaluate bond and anchorage performance in accordance with GB 50367 and related concrete design codes. Specimens are usually cast in concrete blocks or slabs with the embedded length controlled so that bond failure occurs before the steel yields.

Mechanical Fasteners and Inserted Components

Wood screws, self-tapping and self-drilling screws, lag bolts, rivets, threaded inserts in plastics and light alloys, and press-fit or overmolded inserts are tested for withdrawal capacity using the same pull-out principle, with fixtures adapted to the product geometry.

Testing Items

A complete pull-out strength evaluation typically reports the following items:

• Ultimate pull-out load, expressed in kilonewtons (kN), and the corresponding normalized capacity per unit embedment.
• Average bond stress at failure, calculated as the ultimate load divided by the bonded surface area, that is, bond stress equals the pull-out load divided by the product of pi, the bar or anchor diameter, and the effective embedment depth.
• Displacement at ultimate load and displacement at specified service or proof load levels.
• Proof-load verification result for non-destructive in-situ testing, including hold-time behavior and residual slip.
• Failure mode classification, such as steel rupture, anchor pull-out, pull-through, concrete cone breakout, splitting of the base material, or bond interface failure.
• Statistical evaluation of results, including the characteristic resistance derived as a five percent fractile with ninety percent confidence in accordance with EN 1990 and EN 1992-4.
• Stiffness or slip modulus taken from the initial slope of the load-displacement curve.

Testing Methods

Specimen Preparation and Conditioning

Concrete and masonry base materials are normally cured for at least 28 days, and their compressive strength is verified by testing companion cubes or cylinders in accordance with ASTM C39 or EN 12390-3. Holes are drilled, cleaned, and prepared following the anchor manufacturer's installation instructions, and installation torque or displacement is recorded. For adhesive anchors, specimens are conditioned through the full gel and cure time before loading, and testing temperature is recorded because polymer bond strength is temperature dependent.

Direct Tensile Pull-Out Test

The specimen is mounted in a calibrated hydraulic or servo-electric pull machine, and a steadily increasing axial load is applied until failure. In accordance with ASTM E488, the loading rate is controlled so that failure occurs within approximately one to three minutes, ensuring quasi-static conditions. Load is measured with a calibrated load cell meeting the accuracy requirements of ISO 7500-1, while displacement is measured with a transducer referenced to the undisturbed concrete surface.

Confined Versus Unconfined Test Setups

In an unconfined setup, the reaction frame bears on the base material outside the expected failure cone, typically at a support distance of at least twice the effective embedment depth, which allows the full breakout cone to form. In a confined setup, the reaction ring sits close to the anchor and suppresses cone breakout, isolating bond or pull-out failure. The choice of configuration depends on whether the test is intended to determine steel or bond capacity or the concrete breakout resistance used in design.

Proof-Load Verification Test

For in-situ quality control, portable hydraulic pull-out testers apply a specified verification load, commonly between 1.0 and 1.5 times the design load, and hold it for one to two minutes. If the anchor sustains the load without pull-out, excessive displacement, or load relaxation beyond the specified limit, the installation is accepted. This method is widely used for sampling installed anchors on completed structures.

Rebar Bond Pull-Out Test

Reinforcing bar specimens are pulled while slip is monitored at both the loaded end and the free end. The bonded length is deliberately limited so that the average bond stress reaches failure before the bar yields. The average bond strength is calculated from the maximum load and the nominal bonded area, and results are compared with the design bond values required by the applicable structural code.

Data Evaluation and Reporting

Individual ultimate loads are recorded for each replicate, and a minimum of five specimens per configuration is generally required for statistical validity, with larger populations used for product qualification programs. Results are normalized to the actual measured concrete strength where required, the characteristic value is calculated using recognized fractile methods, and every failure mode is documented photographically. The final report includes the equipment calibration records, specimen details, load-displacement curves, and a clear statement of conformity with the specified acceptance criteria.

Applications

Pull-out strength testing supports a broad range of engineering and manufacturing activities:

• Qualification and design verification of post-installed anchors for façade systems, curtain walls, machinery foundations, and steel-to-concrete connections.
• Acceptance testing of post-installed rebar used in structural strengthening and extension projects.
• In-situ quality control of installed anchors on construction sites using portable proof-load equipment.
• Quality assurance for fasteners and threaded inserts in furniture, automotive, appliance, and electronics assemblies.
• Research and development of new anchoring adhesives, undercut systems, and fastener geometries.
• Failure analysis and technical dispute resolution where installed anchorage performance is questioned.

FAQ

What is the difference between destructive and non-destructive pull-out testing?

Destructive testing loads the anchor to failure to determine the ultimate capacity and failure mode. Non-destructive proof-load testing applies only a predefined verification load, and the anchor is accepted if it holds the load within the allowable displacement and relaxation limits, leaving it serviceable afterward.

How many specimens are required for a valid pull-out test series?

Most laboratory programs require a minimum of five identical specimens per configuration to allow statistical evaluation. Product qualification schemes under EN 1992-4 or ASTM E488 may require larger series tested across multiple concrete strengths, crack conditions, and embedment depths.

Which failure modes are considered acceptable?

Acceptable modes depend on the design intent. Ductile steel rupture is often preferred for seismic applications, while concrete cone breakout, pull-out, and bond failure may be acceptable if the measured characteristic resistance exceeds the required design value. Splitting failure usually indicates inadequate edge distance or spacing and is commonly treated as non-conforming.

How is the loading rate controlled?

The load is increased smoothly under displacement or load control so that the time to failure falls within the one-to-three-minute window specified by ASTM E488, avoiding both dynamic effects and long-duration creep influences. Proof-load tests additionally include a sustained hold period at the target load.

Can pull-out testing be performed on existing structures?

Yes. Portable hydraulic pull-out testers allow proof-load verification on installed anchors, post-installed rebar, and similar connections in completed buildings. Test locations should be selected to avoid compromising structural integrity, and any anchors loaded to failure must be replaced or re-justified by the responsible engineer.