Designing structural anchorage to concrete

January 1, 2010

Photo ©

By J. Bret Turley, PE
In the last issue of Construction Canada[2], this author explored the various designs of structural anchors intended for masonry assemblies. Given how the methodology for designing these structural anchorages changed significantly in 2004, a closer look at similar products for use with concrete is warranted.

Annex D of Canadian Standards Association (CSA) A23.3-04, Design of Concrete Structures, introduces a new and comprehensive limit states design (LSD) procedure for determining factored tension and shear resistance of both cast-in-place (CIP) anchors and pre-qualified post-installed mechanical anchors installed in cracked and uncracked concrete. It now also specifies the test standard by which post-installed mechanical anchors are to be pre-qualified for use under LSD methodology.

Annex D is an informative, rather than mandatory, part of the standard. However, as it represents the state of the art relative to the design of structural anchorage to concrete, it is written in ‘mandatory’ language to facilitate its formal adoption by design professionals or regulatory authorities. CSA A23.3-04 is referenced by the 2005 National Building Code of Canada (NBC), which forms the basis of most provincial building codes. Apart from Annex D, there are no other standards referenced by NBC or provincial codes that address the design of structural anchorage to concrete.

CSA A23.3-04 was amended in August 2009 to include revisions to Annex D pertinent to the seismic design of anchors. While the annex does not currently address post-installed adhesive anchors, there is a CSA subcommittee working to develop the design and test provisions necessary to incorporate these product types.

New design methodology
The design of anchors under Annex D’s limit states design method is either based on:
• calculation using design models that result in the prediction of resistance in substantial agreement with the results of comprehensive tests; or
• test evaluation using the five per cent fractile values (or characteristic values) of the test results to determine the factored resistance for seven possible failure modes.
The latter approach is predominately used.

There are four possible failure modes for tension:
• steel fracture;
• concrete breakout;
• anchor pullout or pull-through; and
• concrete side-face blowout (this is applicable for CIP headed anchors only).

Similarly, there are three possible failure modes for shear:
• steel fracture;
• concrete breakout; and
• concrete pry-out.

Failure modes for anchors under tension loading. Images courtesy Canadian Standards Association
Failure modes for anchors under shear loading.

Figures 1 and 2 (left) depict the various possible failure modes for tension and shear, respectively. The proper resistance modification factors are determined giving consideration to a few factors:
• whether the steel is ductile or brittle;
• sensitivity to installation effort and overall reliability (e.g. anchor category applicable to post-installed mechanical anchors only); and
• whether supplementary reinforcement is used.

In calculating concrete breakout in tension or shear, modification factors adjust the factored resistance. This is to account for variables such as:
• absence of concrete cracking;
• premature splitting;
• groups of anchors loaded eccentrically; and
• close edge effects.

Currently, pullout or pull-through for post-installed mechanical anchors in tension must be determined by testing and using the five per cent fractile value in the calculation of the factored resistance. The lowest factored resistance for tension and shear are determined by considering all applicable failure modes and by comparing the greatest factored tension and shear force, considering all applicable load combinations. The factored resistance should be greater than or equal to the factored forces.

In the case where factored tension and shear forces act concurrently on an anchor (or group of anchors), an interaction relationship must also be satisfied. When post-installed mechanical anchors are specified to resist seismic forces, they must pass the simulated earthquake testing program.

Software automates design of post-installed anchors under Annex D. Image courtesy Simpson Strong-Tie Anchor Systems

Given the calculation-intensive nature of this methodology, and the need to design multiple anchorage to concrete connections on any given project, most design professionals either develop spreadsheets to ‘automate’ the process or rely on software. Figure 3 (right) is an example of a graphic user interface developed for the design of post-installed anchors under Annex D.

The need for a test standard
In the 1994 and 2000 editions of CSA A23.3, Annex D was referred to as ‘Appendix D.’ It specified the need to both qualify and quantify performance of post-installed anchors installed in the tension zone of a concrete member (if that was the intended use). However, at the time, there were no standards available to address the testing and assessment of post-installed anchors in cracked concrete.

Consequently, design professionals were directed to either refer to the post-installed anchor manufacturer’s test data or develop and conduct testing programs to verify adequacy of the proposed post-installed anchor type for an intended application.

The need for a test standard was fulfilled by the 2001 publication of American Concrete Institute (ACI) 355.2-01, Evaluating the Performance of Post-installed Mechanical Anchors in Concrete. ACI 355.2 provides for the testing and assessment of post-installed mechanical anchors for use in concrete that is either both cracked and uncracked or solely uncracked.

Load-displacement behaviour requirements of post-installed mechanical anchors. Image courtesy American Concrete Institute

The 2004 edition of CSA A23.3’s Annex D specifically references ACI 355.2 for the prequalification of post-installed mechanical anchors for use with LSD methodology. Under ACI 355.2, there are four types of tests performed: identification, reference, reliability, and service condition. The last three assess performance.

Reference tests
Reference tests establish both a baseline for performance and the anchor category. Reference static tension tests are conducted in low- and high-strength uncracked or cracked concrete. The crack width in the reference tests is a minimum of 0.3 mm (0.01 in.). The results of the reference tests are used to determine the effectiveness factor (k), which helps determine concrete breakout strength.

Reliability tests
Reliability tests assess anchor behaviour and performance under normal and adverse conditions, both during installation and in service. Figure 4 (right) depicts acceptable and unacceptable load-displacement curves that represent part of the evaluation process of post-installed mechanical anchors.

These analyses include static tension tests in cracked or uncracked concrete with reduced installation effort or installations in small- or large-hole diameters. The crack width in the reliability tests is 0.3 mm (0.01 in.) for the reduced installation effort tests; for the sensitivity to hole diameter tests it is 0.5 mm (0.02 in.). Procedures are conducted in low- or high-strength concrete as specified in the criteria. One of the more demanding reliability tests requires the anchor be subjected to a sustained tension load, with the crack width cycled between 0.1 mm (0.004 in.) and 0.3 mm for 1000 cycles.

These two wedge-type expansion anchors look similar, but the one on the right—a redesign—is the only one suitable for cracked concrete. The anchor on the left is unacceptable.

The results of the reference and reliability tests establish the anchor category for each post-installed anchor by diameter. This indicates an anchor’s sensitivity to installation and reliability, which is then used to determine the appropriate resistance modification factor.

Service condition tests
Service condition tests are conducted to establish the data that then  predicts anchor performance under service conditions. The service condition tests include:
• single-anchor installations at close proximity to a corner;
• minimum edge distance and spacing to preclude splitting;
• static shear in cracked concrete;
• simulated seismic tension in cracked concrete; and
• simulated seismic shear in cracked concrete.
Crack width in the simulated seismic tension and shear tests is 0.5 mm (0.02 in.). All service condition tests are run in low-strength concrete.

Anchors meeting Annex D
The change to Annex D has prompted development of a new generation of post-installed anchors. Figure 5 shows two types of post-installed mechanical anchors pre-qualified for use in both cracked and uncracked concrete. In some instances, products have been redesigned to satisfy the new criteria.

In Figure 6, two wedge-type expansion anchors appear similar in design. The anchor on the left is not qualified for use in cracked concrete. The anchor on the right (a redesign of the product on the left) is qualified for use in cracked concrete due to a change in materials and expansion mechanism design.

From the previous discussion of the testing requirements, it is apparent post-installed mechanical anchors are subject to a much more demanding test regimen and assessment requirements for use under Annex D. Consequently, not all post-installed anchor products are suitable for use in cracked or uncracked concrete. It is therefore important the design professional specify only those products pre-qualified under ACI 355.2 when the anchorage to concrete connection has been designed under the Annex D methodology.

J. Bret Turley, PE, is the manager of technical services for Simpson Strong-Tie Anchor Systems. He has been active in the anchoring and fastening industry for more than 20 years. Turley is the secretary of the American Concrete Institute (ACI) Committee 355, Anchorage to Concrete, and is also a member of the committees 349-C, Nuclear Structures–Anchorage, and 408, Development & Splicing of Deformed Bars. He serves on the Fédération Internationale du Béton Special Activities Group 4 on Fastenings to Structural Concrete and Masonry. Turley may be contacted at

  1. [Image]:
  2. Construction Canada:
  3. [Image]:
  4. [Image]:
  5. [Image]:
  6. [Image]:
  7. [Image]:

Source URL: