Long-Term Performance of Epoxy Adhesive Anchor Systems (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

108 This section presents the proposed specifications for AASHTO relating to material and testing, design, construc- tion, and quality assurance for adhesive anchor systems in concrete. A flowchart is provided in Appendix N to better describe how the various ICC-ES, ACI, and AASHTO stan- dards and specifications fit together and interact. Material and Testing ACI 355.4 provides testing and material specifications for adhesive anchors in concrete. This specification includes a large battery of tests and criteria for adhesive anchors in cracked and uncracked concrete. ACI 355.4 has evolved from ICC-ES (1995) AC58, which initially established the acceptance criteria for adhesive anchors in concrete and masonry and eventually from ICC-ES (2008) AC308 for adhesive anchors in concrete alone. Both ICC-ES documents are based on testing methods presented in ASTM E488 and ASTM E1512. ACI 355.4 includes the following mandatory tests for systems approved for use in both cracked and uncracked concrete: • Reliability Tests: – Sensitivity to hole cleaning, dry concrete; – Sensitivity to hole cleaning, water-saturated concrete; – Sensitivity to mixing effort; – Sensitivity to installation, water-saturated concrete; – Sensitivity to crack width in low-strength concrete; – Sensitivity to crack width in high-strength concrete; – Sensitivity to crack width cycling; – Sensitivity to freezing/thawing conditions; – Sensitivity to sustained load; and – Torque test. • Service-condition Tests: – Tension in low-strength concrete, – Tension in high-strength concrete, – Tension in low-strength cracked concrete, – Tension in high-strength cracked concrete, – Tension at elevated temperatures, – Curing time at standard installation temperature, – Resistance to alkalinity, – Edge distance in corner condition to develop full capacity, – Minimum spacing and edge distance to preclude splitting, – Shear capacity of anchor element having a non-uniform cross section, and – Round robin tests for regional concrete variation. ACI 355.4 includes the following optional tests. Anchors not tested for these optional parameters will have limitations placed on their use: • Reliability Tests: – Sensitivity to hole cleaning, water-filled hole; – Sensitivity to hole cleaning, submerged concrete; – Sensitivity to installation, water-filled hole; – Sensitivity to installation, submerged concrete; and – Sensitivity to installation direction. • Service-condition Tests: – Tension at decreased installation temperature, – Resistance to sulfur, – Seismic tension, – Seismic shear, and – Member minimum thickness. Testing Specifications (Interim Proposal) It is proposed that, in the short term, AASHTO adopt the ACI 355.4 testing program for acceptance of adhesive anchors in transportation projects and not develop its own program for several reasons: • ACI 355.4 has undergone the extensive ANSI consensus review process and has been adopted by other code and standard organizations. C H A P T E R 4 Proposed AAshto specifications

109 and residual strength criteria are met at 75 years at standard temperature and 15 years at elevated temperature. However, in order to avoid additional testing as discussed above, the SvTTF relationship can be used to determine a reasonable sustained load reduction factor for AASHTO design applica- tions. Table 45 shows that all of the six individual and three combined baseline SvTTF curves project an average drop of 1%MSL between 10 years and 15 years and a drop of 2% between 10 years and 20 years. It is important to realize that the 55% stress level at which manufacturers conduct their sustained load tests is not actually 55% of the short-term strength. It is common that sustained load tests at 55% of the actual short-term strength will not pass the displacement criterion of the test program. Therefore, manufacturers often downgrade their short-term strength in order to pass the sustained load test. This down- graded short-term strength becomes the new short-term bond • This testing program has a very significant cost per product line and it would be prohibitive to require manufacturers to evaluate their products under another testing program for use in transportation projects. • There are several products on the market that have under- gone the extensive testing program specified in ICC-ES AC308, which are currently being evaluated under the slightly modified provisions of ACI 355.4. AASHTO can immediately incorporate these approved products in trans- portation projects. • The ACI 355.4 testing method results in a conservative and reasonable reduction factor for use in adhesive anchor design for sustained load as discussed below. The sustained loading test program in ACI 355.4 subjects an anchor to 55% of the mean static load at standard temperature and at the long-term elevated temperature for 1,000 hours. The displacements from the last 20 days are projected based on a logarithmic trend. The projected displacements at 50 years for the standard temperature test and 10 years for the long-term elevated temperature test must be less than the displacement at loss of adhesion (Dlim) for their respective short-term tests (Figure 114). Additionally, the residual strength must be at least 90% of the short-term test strength. Anchors that pass these cri- teria are acceptable for use in concrete structures up to a lifetime of 50 years. The 55% stress level coupled with the displacement criteria results in a 0.55 reduction factor in the ACI 318-11 design provisions for adhesive anchors in sustained load applications. As AASHTO stipulates structure lifetimes of 75 years as compared to 50 in ACI, a different (lower) modification factor must be specified. One option is to require testing at a lower stress level that will guarantee that the displacement Table 45. Projected stress level at 10 years and drop in stress level for various structure lifetimes from the baseline SvTTF curves. 10 years Drop Is Drop Is Baseline Test (87,600 hrs) 10 yrs to 15 yrs 10 yrs to 20 yrs 01-A 33 1 3 01-B 42 2 2 01-C 54 1 2 02-A 35 1 1 02-B 56 1 2 02-C 56 1 1 A-combined 33 1 2 B-combined 50 1 2 C-combined 48 1 2 %MSL from %MSL from Figure 114. Current ACI 355.4 sustained load test projection.

110 protocol for use in sustained load applications in transpor- tation structures or (2) work with ACI in transitioning the sustained load testing in ACI 355.4 toward an SvTTF testing program. It should be noted that there was a large scatter on time to failure for the same stress level, indicating that several tests will need to be required. It was shown in Table 38 and Figure 55 that the average creep displacement is 1.3 to 2.2 times the average short-term peak displacement for the three adhesives tested and in Table 39 was shown to be 1.4 to 3.0 times the displacement at loss of adhesion per ACI 355.4. It would be reasonable then to project the displacements for tests that have not failed to a lower bound value of the creep displacements of tests that have failed as opposed to using the displacement limit based on the short- term displacements. However, if the creep displacement data is normalized by the average creep displacement for each adhesive and pooled together, the resulting lower bound values (using a 5% frac- tile) of the creep displacement in terms of the short-term displacement for each adhesive is presented in Table 46. The normalized lower bound values for adhesive A are 1 for both approaches and those for adhesives B and C are between 1.2 and 1.8. Based on this analysis, it appears that the ACI 355.4 projection of sustained load test data to the average short-term displacement is a rational approach. The current projection method does not take into account tertiary creep or rupture but assumes a limit on displacements under sustained load. For example, the three repetitions of the US adhesive C baseline at 65% resulted in three different displacement versus time responses (Figure 115). Repetition 8 exhibited a standard tertiary creep region with rupture at 370 hours. Both repetitions 7 and 9 exhibited a significant reduction in displacement rate between 1,000 and 2,000 hours, possibly indicating that the anchors would cease to displace and sustain the load indefinitely. Repetition 7 has essentially ceased to displace and was terminated at 15,000 hours due to labora- tory logistics. However, repetition 9 had a sudden increase in displacement rate after 4,000 hours and ruptured. These three repetitions at the same stress level indicate that a projection strength that is lowered again to account for other parameters with the end result being that the final published value for bond stress is significantly lower than the actual short-term bond strength. As the test data is protected by the manufac- turers, based on our calculations we estimate that the down- graded short-term bond strength is in the range of 65%MSL to 75%MSL and the subsequent sustained load tests are con- ducted in the range of 35%MSL to 40%MSL. Based on these estimates, the projected stress level to cause failure at 15 years is approximately 1%MSL less than the stress to cause failure at 10 years and 2% less at 20 years. Therefore, it seems reasonable to specify a reduction factor for AASHTO at 15 years no less than 1%MSL below the 55% stress level at 10 years. In other words, a reduction factor for design of adhesive anchors under sustained load for AASHTO should not be greater than 0.54. To allow for additional con- servatism, it is suggested that a reduction factor of 0.50 be used for lifetimes of 75 and 100 years. A modification to ACI 355.4 §8.5.2.3 has been included to test anchors at temperature category B for long-term tem- peratures above 120°F. Additionally, it is proposed that AASHTO require a series of tests to determine the alpha-setup factor for the relation- ship between unconfined tests to confined tests as opposed to using the default value of 0.75 specified in ACI 355.4 §10.4.5.1. Finally, a modification to ACI 355.4 §7.9 includes a quali- fication test for evaluating the effect of pausing during the installation of the adhesive. Testing Specifications (Proposal to Include SvTTF) It is proposed that sustained load testing for adhesive anchors eventually transition from the current pass/fail criteria towards an SvTTF approach. AASHTO TP 84-10 was recently created to provide a framework for SvTTF testing for adhesive anchors. AASHTO can either (1) choose to require manufac- turers to qualify their products under an AASHTO SvTTF Adhesive Lower Bound Creep Displacement/ Average Short-Term Displacement Lower Bound Creep Displacement/ Average Limiting Displacement A 1.0 1.0 B 1.2 1.8 C 1.3 1.7 Notes: Based on a K value (5% probability of nonexceedence with a confidence of 90%) of 1.64. Characteristic value for adhesive A used the COV for adhesive A. Characteristic value for adhesives B & C used the pooled COV from all adhesives. Table 46. Characteristic creep displacement for the three adhesives used in this project.

111 the sustained load tests that failed]. This projection is to be based on the last 20 days of data (minimum 20 data points). Furthermore, the manufacturer can opt to continue sus- tained load testing and update its SvTTF curve in the future providing a higher published sustained load strength. A proposed Standard Method of Test for Adhesive Anchors in Concrete, which references ACI 355.4-11 and AASHTO TP 84-10, is included in Appendix O. Material Specifications A proposed Standard Specification for Adhesive Anchors in Concrete is included in Appendix P. Design Design Specifications The consensus between the NCHRP panel and the research- ers was to provide a simplified AASHTO design specification intended to cover the majority of adhesive anchor systems under tensile loading with a few simplifying assumptions. For anchor applications that fall outside the limitations of the AASHTO design specification, the designer is encouraged to refer to ACI 318-11 Appendix D due to the overwhelming time and effort invested in its development. There is precedent from AASHTO to reference ACI documents in regards to anchors as AASHTO (2010b) LRFD Bridge Design Specifications §C6.13.2.12 and §C14.8.3.1 currently refers to ACI 318-05 Appendix D for “the global design of anchorage to concrete.” method assuming continually increasing displacements would not identify these three responses. However, as the displace- ments seen in these tests are well above the limiting displace- ment determined by ACI 355.4 the current projection method is conservative. A benefit of the SvTTF approach is that manufacturers will not have to downgrade their published short-term strength, but can publish their actual short-term strength and a sustained load strength for particular lifetimes. Based on this research project, a few modifications are proposed for TP 84-10. Section 9.4.2 of TP 84-10 requires two stress levels for sustained load testing (70%–80%) and (60%–70%) of the MSL. Based on the early failures in the sustained load tests and that the regression analysis of the baselines indicates an approximate 80% strength at 5 minutes, it is proposed that the highest sustained load test should be conducted at 70%MSL. Based on the scatter, two other stress levels should be tested at 60% and 50%. Furthermore, as dis- cussed previously, results from short-term tests should not be included in the SvTTF relationship. It is further proposed that tests should be conducted to failure and the failure times plotted on the SvTTF curve. For those tests that do not fail within a reasonable timeframe, the manufacturer can include these tests on the SvTTF curve at the larger of the following two times: • The current test duration. • Projected time to reach the limiting displacement from the short-term tests [which was shown to be essentially the lower bound (5% fractile) of the creep displacement from Figure 115. University of Stuttgart baseline tests for adhesive C at 65%MSL.

112 The following limitations are included in the AASHTO design specification. The commentary in the design specifi- cation discusses the justification behind their inclusion. • Products shall be qualified for use in cracked concrete in accordance with ACI 355.4. • The effective depth of embedment, hef, must not be less than 4da, 15/8”, or the minimum stated in the manufacturer’s printed installation instructions (MPII). • The effective depth of embedment, hef, must be less than or equal to 20da or the maximum stated in the MPII, which- ever is less. • Edge distance, c, from the center of the anchor to the nearest edge of concrete must not be less than the larger of 6da or the minimum stated in the MPII. • Anchors must be installed in holes drilled with a manu- facturer approved rotary impact drill or rock drill unless permitted by MPII. • Concrete must be normal weight concrete. • The concrete member is considered cracked with normal temperature and shrinkage cracks and with minimum reinforcement. • The concrete at time of installation shall have a minimum temperature of 50°F (10°C) or that stated in the MPII, whichever is greater. • Concrete at time of installation shall have a minimum age of 21 days and a minimum compressive strength of 2,500 psi. • The tensile loading on the group of anchors must be applied centrically to the anchor group. • Anchors must not be subjected to seismic loads. The proposed AASHTO Design Specification for Adhesive Anchors is included in Appendix Q. Design Guidelines A proposed AASHTO Design Guideline for Adhesive Anchors is provided in Appendix R, which provides a brief orientation on adhesive anchor design and refers the user to ACI 318-11 Appendix D for the extensive commentary on the design of adhesive anchors in concrete. Additionally, two calculation examples (for a single anchor in tension and for a group of anchors in tension) are provided using the proposed AASHTO design specifications for adhesive anchors. Construction and Quality Assurance Quality Assurance Guidelines A Quality Assurance Guideline has been drafted and is included in Appendix S. It is based on information from ACI 355.4, CRSI (2011), Wollmershauser and Mattis (2008), Mattis and Silva (2011), and various manufacturer printed installation instructions (MPII). This document is intended to serve as general information for the installer and inspector. It discusses the various types of adhesives, installation equip ment, and storage and handling suggestions. It is also intended to provide the “why” behind the many common installation instructions found in most MPII. As this is a guideline, it is not written in mandatory language and often refers the reader to the MPII for all storage, handling, and installation procedures. Construction Specifications Section 29 of the AASHTO 2010 LRFD Bridge Construction Specifications currently addresses embedment anchors and includes references to bonded anchors systems and adhesive anchors. Also included are general references to qualification by universal tests standards and certifications by an engineer. ACI 349 Appendix B was previously referenced for embed- ment anchor details. Section 29 provided a solid framework to include more specific construction specifications related to adhesive anchor systems. The following briefly discusses the proposed changes to AASHTO 2010 LRFD Bridge Construction Specifications Sec- tion 29 and are included in underline/strikeout format in Appendix T. 29.2-PREQUALIFICATION • Includes references to ACI 355.4 for post-installed adhe- sive anchors. 29.3-MATERIALS • Adopts the ACI 355.4 definition of acceptable adhesive products. The current definition could inadvertently restrict the use of some products that meet the assessment criteria of ACI 355.4. • Requires the use only of products that meet the assessment criteria of ACI 355.4. • Provides a definition of adhesive anchor systems. • Prohibits the use of bulk adhesive mixed in open containers without automatically controlled metering and mixing of components. • Includes a statement on storage and discarding of expired product. 29.4-CONSTRUCTION METHODS • Includes references to the MPII for adhesive anchors related to installation procedures. • Includes limitation from ACI 355.4 on the minimum age (21 days) and compressive strength (2,500 psi) of concrete for installation of adhesive anchors.

113 • Requires anchors for sustained load applications to not be loaded or torqued until 24 hours after the manufacturer’s minimum cure time. REFERENCES • Updates the references section to include ACI 318 and ACI 355.4. Summary This chapter presented the rationale behind the development of proposed standards and specifications for AASHTO per- taining to the testing, design, construction, and inspection of adhesive anchors in concrete. The proposed drafts are included in Appendixes O through T. • Includes requirement that adhesive anchors installed in horizontal or upwardly inclined holes shall be installed by ACI/CRSI certified personnel. • Article C29.4: Discusses MPII, installation limitations, and the ACI/CRSI Adhesive Anchor Installer Certification program. 29.5-INSPECTION AND TESTING • Requires continuous inspection on adhesive anchors installed in horizontal and upwardly inclined holes. • Requires adhesive anchors exposed to in-service tem- peratures ≥120°F (49°C) to be tested and qualified under Temperature Category B at a temperature equal to or greater than the highest in-service temperature. • Includes requirements for proof testing of adhesive anchors.