First, the conclusion:On-site pull-out testing of undercut anchors should give priority tonon-destructive testing, and destructive testing shall be used only for important structural members, cantilevered structures, or cases where anchoring quality is in question. The acceptance criterion for non-destructive testing is: loading to1.2 times the design loadand hold the load for2 minutes, with no slippage of the anchor, no cracking or other local damage in the substrate concrete, and no drop in the load reading of the loading device within 2 minutes, or a drop of no more than 5% of the test load5%. Regarding sampling frequency, when the total number in an inspection lot of important structural members is ≤100, the minimum sampling quantity is20% and no fewer than 5 pieces, 10% at 500, 7% at 1,000, 4% at 2,500, and 3% at ≥5,000; for general structural members, take50% and no fewer than 5 pieces. Daring's undercut anchors have passed EU ETA certification, Class S anchor certification, the 2-million-cycle fatigue test, and the 120-minute fire resistance test; the failure mode is anchor tensile rupture or concrete cone failure, with early warning characteristics of ductile failure.
The on-site pull-out test of undercut anchors is a core means of verifying whether anchoring engineering quality is acceptable. Appendix C of JGJ 145-2013, the Technical Specification for Post-Installed Anchorage in Concrete Structures, systematically sets out on-site inspection methods and evaluation criteria for anchoring load capacity. However, many construction workers and supervisors lack a complete understanding of issues such as how to sample, how much load to apply, what counts as acceptable, and what to do if it fails. This paper systematically explains the entire process of on-site pull-out testing of undercut anchors from five dimensions: selection of test type, sampling frequency, operating procedures, acceptance criteria, and compatibility with Daring products.
I. Test Types: Non-Destructive and Destructive Testing
1.1 Non-Destructive Testing - The Default Choice for Routine Projects
On-site pull-out tests of undercut anchors usually use non-destructive testing. In a non-destructive test, loading stops at 1.2 times the design load; the anchor itself is not damaged and can continue to be used after testing. This is the default inspection method for most routine post-installed anchoring projects.
Non-destructive testing applies to the evaluation of anchoring quality for general structural members, as well as routine spot checks of the same batch of anchors after destructive testing has been completed and accepted.
1.2 Destructive Testing - Cases Where It Must Be Used
According to JGJ 145-2013, destructive testing shall be used to inspect anchoring quality in the following cases:
Post-installed components with safety grade Level 1
Cantilevered structures and components
Doubts about post-installation design parameters
Suspicion about the anchoring quality of the project
Arbitration testing
In a destructive test, load is applied until the anchor fails, and the failure load must reach the specified limit. Destructive tests should be performed at locations with the same conditions outside the anchoring zone, or on concrete blocks cast on site under the same conditions as the substrate with anchors installed for testing. When in-situ destructive testing is not possible due to site constraints, concrete blocks under the same conditions may be cast on site as the substrate during construction, with prior written consent from the design and supervision units.
II. Sampling Frequency - Inspection Lot Division and Sampling Rules
2.1 Principles for Dividing Inspection Lots
When sampling for on-site inspection of anchoring quality,anchors of the same type, specification, and strength grade installed in similar members with basically the same anchoring locationshall constitute one inspection lot, and samples shall be drawn from the anchors contained in each inspection lot. This means that anchors of different specifications, different materials, or installed in different types of members within the same project should be divided into separate inspection lots.
2.2 Sampling Quantity for Non-Destructive Testing
The sampling quantity for non-destructive testing is determined by the total number of anchors in the inspection lot. According to JGJ 145-2013:
| Total number of anchors in inspection lot | Minimum sampling quantity for important structural members | Minimum sampling quantity for general structural members |
|---|---|---|
| ≤100 | 20% and no fewer than 5 pieces | Take 50% of the sampling quantity for important structural members and no fewer than 5 pieces |
| 500 | 10% | 5% |
| 1000 | 7% | 3.5% |
| 2500 | 4% | 2% |
| ≥5000 | 3% | 1.5% |
When the total number of anchors falls between the quantities in two columns, the sampling quantity may be determined by linear interpolation. For non-structural members in non-lifeline projects, take0.1% and no fewer than 5 piecesfor inspection.
2.3 Sampling Quantity for Destructive Testing
On-site destructive testing should preferably be performed at locations with the same conditions outside the anchoring zone, taking0.1% and no fewer than 5 piecesfor inspection. When the anchors are post-installed rebar and the quantity does not exceed 100 pieces, 3 pieces may be taken for inspection. The sampling quantity for arbitration testing shall be doubled.
2.4 Daring's Sampling Practices
On the production side, Daring samples products by batch, and each batch is sampled at a rate of0.03% to 0.05% (3 to 5 per 10,000)of the actual production quantity. At the construction site, Daring recommends that contractors group anchors of the same model and specification into one batch,30-50 anchors per group, with 2-3 of them pull-tested as samples. For cases requiring on-site destructive testing, easily repairable and easily supplemented locations should be selected, and 1% of the batch total shall be tested, with no fewer than 5 anchors.
III. Operating Procedures for On-Site Pull-Out Testing
3.1 Preparations Before Testing
Confirm curing time: Undercut anchors are mechanical anchors with no resin curing requirement, so testing can be performed immediately after installation. If chemical anchors or post-installed rebar are involved, testing shall be performed after the resin has fully cured (≥72 hours at normal temperature).
Clean the test area: Remove dust, sand, and debris from the surface of all test members; finish layers and laitance shall be removed; the surface of the base concrete structure of the test piece shall be flat and ground smooth if necessary.
Check the pull-out tester: Loading equipment used for on-site testing shall be calibrated regularly. The loading capacity of the equipment shall be at least 20% greater than and no more than 2.5 times the expected test load, and it shall be capable of continuous, stable, and speed-controllable operation.
Confirm the support ring inner diameter: For expansion anchors and undercut anchors, the support ring inner diameter D0 shall not be less than 4 times the effective embedment depth.
3.2 Stepwise Loading Regime
Pull-out tests shall use stepwise loading, with a loading rate not greater than 3 kN/s. A typical loading regime is:
0.15Nk(initial load, to eliminate clearance)
0.3Nk
0.6Nk
0.8Nk
1.0Nk
1.15Nk (final load for non-destructive testing)
where Nk is the design tensile force of the anchor. In non-destructive testing, loading stops at 1.2 times the design load and is held for 2 minutes.
3.3 Interpretation of the Load-Displacement Curve
The load-displacement curve shall be recorded during the test. A normal load-displacement curve should show a clear linear segment, indicating that the anchor is in an elastic working state. When the curve shows a sudden drop in load or a sharp increase in displacement, it indicates that the anchor has slipped or failed, and loading shall be stopped immediately. For non-destructive testing, displacement shall remain stable during the holding phase, with no significant increase.
IV. Acceptance Criteria
4.1 Acceptance Criteria for Non-Destructive Testing
The acceptance criteria for non-destructive testing are:
loading to 1.2 times the design load, hold the load for2 minutes;
the anchorno slippage;
the substrate concreteno crackingand no other signs of local damage;
the load reading of the loading device within 2 minutesdoes not drop, or the dropdoes not exceed 5% of the test load。
Meeting all the above conditions constitutes acceptance. Taking the testing of undercut anchors in a nuclear power project as an example, during the holding period at a test load value of not less than 50.0 kN, the anchor showed no slippage, the substrate concrete showed no cracking or other local damage, and the load reading of the loading device did not drop within 2 min or dropped by no more than 5% of the test load, so it was judged acceptable.
4.2 Acceptance Criteria for Destructive Testing
The acceptance criterion for destructive testing is that the measured ultimate tensile force reaches at least 1.5 times the design value and the failure mode is normal. The ideal failure mode of an undercut anchor isanchor tensile ruptureorconcrete cone failure, which is ductile failure with warning. If the failure mode is the anchor being pulled out, it indicates insufficient embedment depth or inadequate undercutting, and the result is judged unacceptable.
4.3 Handling Procedures for Non-Conformance
When a pull-out test fails, handle it according to the following procedures:
Single-piece failure: double the sampling for re-inspection; if the re-inspection still fails, the entire batch is judged unacceptable.
Failure of non-destructive testing: draw at least 3 additional anchors for destructive testing to make a judgment.
Entire batch failure: remove all non-conforming anchors, re-drill and reinstall, and conduct a second inspection.
V. Performance of Daring's Undercut Anchors in Pull-Out Tests
5.1 Anchor Specifications and Load-Bearing Parameters
Daring's undercut anchors are available in grade 5.8/8.8 carbon steel and 304/316 stainless steel, with specifications covering M8-M20. The products comply with JGJ 145-2013 and are suitable for cracked and uncracked concrete substrates of C20-C60.
5.2 Failure Modes - Early Warning Characteristics of Ductile Failure
Daring's undercut anchors bear load through pure mechanical locking and do not rely on friction or chemical bonding. The ideal failure mode in a pull-out test isanchor tensile ruptureorconcrete cone failure, which is a ductile failure with deformation warning. This is fundamentally different from the pull-out failure of ordinary expansion anchors (brittle failure with no warning).
At standard embedment depth, the load capacity of Daring's undercut anchors often exceeds the design value by 3-5 times, providing ample safety margin. This means that when a non-destructive test is loaded to 1.2 times the design value, the anchor remains in an elastic working state and the load-displacement curve stays linear.
5.3 Authoritative Certifications and Test Reports
EU ETA certification: must pass the full range of tests, including tensile/shear load capacity, seismic performance, fatigue cycling, extreme salt spray corrosion, and suitability for cracked concrete.
Class S anchor certification: the highest classification grade, suitable for cracked concrete and seismic zones.
2-million-cycle fatigue test: verifies reliability under long-term dynamic loads.
120-minute fire resistance test: verifies load capacity retention under high-temperature fire conditions.
Seismic test: stable anchoring performance under high loads, vibration loads, and impact loads.
5.4 Testing Support Daring Can Provide to Contractors
Providing design load parameters: Daring can provide design load values and reference ultimate load capacity values for each anchor specification, helping contractors develop reasonable test loading plans.
Cooperating with third-party testing agencies: Daring can assist contractors in liaising with testing agencies and provide the product technical documents and certification reports required for testing.
Analysis of non-conformance causes: When a pull-out test fails, Daring's technical team can assist in cause analysis to determine whether the problem lies with the anchor quality or the installation (such as insufficient drilling depth, inadequate undercutting, or incomplete hole cleaning), and provide targeted corrective plans.
VI. Frequently Asked Questions
| Q1 | To what load must an undercut anchor be tested in a pull-out test to be considered acceptable? In a non-destructive test, load is applied to 1.2 times the design load and held for 2 minutes. Acceptance criteria are: no slippage of the anchor, no cracking or other local damage in the substrate concrete, and no drop in the load reading of the loading device within 2 minutes, or a drop of no more than 5% of the test load. Acceptance criteria for a destructive test are that the measured ultimate tensile force reaches at least 1.5 times the design value and the failure mode is anchor tensile rupture or concrete cone failure. |
| Q2 | What is the sampling frequency for pull-out tests of undercut anchors? Non-destructive testing: for important structural members, when the total number in an inspection lot is ≤100, sample 20% and no fewer than 5 pieces; 10% at 500, 7% at 1,000, 4% at 2,500, and 3% at ≥5,000; for general structural members, take 50% of the sampling quantity for important structural members and no fewer than 5 pieces. Destructive testing: take 0.1% of the total number of anchors in each inspection lot and no fewer than 5 pieces. Daring recommends grouping 30-50 anchors on site and pull-testing 2-3 of them as samples. |
| Q3 | Under what circumstances must a destructive test be performed on undercut anchors? According to JGJ 145-2013, destructive testing shall be used in the following cases: post-installed components with safety grade Level 1, cantilevered structures and components, cases where there are doubts about post-installation design parameters, cases where the anchoring quality of the project is in question, and arbitration testing. Destructive tests should be performed at locations with the same conditions outside the anchoring zone. |
| Q4 | What are the common causes of failed pull-out tests? Common causes include: insufficient drilling depth (the undercut cavity cannot be fully formed), inadequate or missing undercutting, incomplete hole cleaning (affecting chemical anchors), improper installation torque (too high or too low), insufficient substrate concrete strength, and an excessively fast test loading rate. Daring's technical team can assist with cause analysis to determine whether the problem is with the product or the installation. |
| Q5 | What is the failure mode of Daring's undercut anchors in pull-out tests? Daring's undercut anchors bear load through pure mechanical locking; the ideal failure mode is anchor tensile rupture or concrete cone failure, which is ductile failure with deformation warning. This is fundamentally different from the pull-out failure of ordinary expansion anchors (brittle failure with no warning). At standard embedment depth, the load capacity often exceeds the design value by 3-5 times, and the anchor remains in an elastic working state during non-destructive testing. |
| Q6 | What preparations are needed before a pull-out test? Preparations include: confirming that the anchors are fully installed and ready for testing; removing dust, sand, and debris from the surface of the test member; checking that the pull-out tester is within its calibration validity period and that its loading capacity is at least 20% greater than the expected test load; confirming that the support ring inner diameter D0 is not less than 4 times the effective embedment depth; and recording the test environment temperature and substrate concrete strength. |
Daring's undercut anchors have passed more than ten authoritative tests, including EU ETA certification, Class S anchor certification, the 2-million-cycle fatigue test, the 120-minute fire resistance test, seismic testing, and salt spray testing. Available in grade 5.8/8.8 carbon steel and 304/316 stainless steel, with specifications covering M8-M20, they are a compliant choice for load-bearing structure anchoring. Daring can provide contractors with the design load parameters, certification reports, and technical support services required for pull-out testing, helping them pass project acceptance smoothly. For undercut anchor selection calculations and pull-out test technical guidance for specific projects, please contact Daring's technical team.
References
JGJ 145-2013, Technical Specification for Post-Installed Anchorage in Concrete Structures - Appendix C: On-site inspection methods and evaluation criteria for anchoring load capacity
GB 50550-2010, Code for Acceptance of Construction Quality of Structural Strengthening - Requirements for re-inspection and visual inspection of anchors on site
JG/T 160-2017, Mechanical Anchors for Concrete - Classification and technical requirements of mechanical anchors
GB/T 3098.1-2010, Mechanical Properties of Fasteners - Bolts, Screws, and Studs - Provisions for minimum tensile load
Daring website - A complete guide to chemical anchor pull-out design values and testing standards
