Full-scale tests of precast pretensioned prestressed double tees with effective depths of 700–980 mm showed that six major design codes (GB 50010-2010, JTG D62-2012, ACI 318-14, AASHTO 2012, fib Model Code 2010 Level I, Eurocode 2) predict shear strengths that deviate substantially and inconsistently from measured capacities, because none contains an explicit size-effect term and none checks for anchorage bond failure, which was observed to govern one test series at a 40% lower load than the companion web-shear failure. The open problem is to establish a predictive shear strength formulation for these members that makes the dependence on member depth explicit, accounts for the effective-prestress profile within the transfer length near the member end, and demarcates the conditions under which strand anchorage bond failure rather than web-shear failure governs the capacity, validated against full-scale experimental data.

Background

A precast prestressed double tee is a standard precast floor or roof member with two vertical ribs (webs) joined by a wide horizontal flange, fabricated by pretensioning: steel strands are tensioned before concrete is cast and released afterwards, so the prestressing force is transferred to the concrete by bond along a transfer length near each member end, over which the effective prestress rises from zero to its full design value. Double tees are produced with thin webs, large depths (here 700–980 mm effective depth, 18 m and 24 m spans), and either a thin cast flange ('untopped') or a thicker flange with cast-in-place topping ('pretopped'); they are widely used in parking structures and long-span buildings.

In shear, such members are designed by semi-empirical code provisions that combine a concrete contribution and a transverse-reinforcement (stirrup) contribution. Several codes base the reinforcement term on a 45° truss model, in which diagonal concrete compression struts inclined at 45° are crossed by vertical stirrups; others (AASHTO, fib Model Code 2010) use the modified compression field theory, a mechanics-based model of cracked concrete in shear. A long-standing experimental fact is the size effect in shear: the nominal shear stress at failure decreases as member depth increases, so equations calibrated mostly on reduced-scale laboratory specimens (in the ACI-DAfStb shear database roughly 79% of specimens with stirrups have effective depths below 400 mm) can be unsafe for full-scale members. None of the six code provisions mentioned above contains an explicit term reducing nominal shear strength with depth for prestressed members, and none treats loss of strand-concrete bond near the member end as a failure mode that can preempt web-shear failure (diagonal-tension cracking of the web).

Zhou, Li and Zhang (2021) tested four full-scale double tees (specimens DT-18, DT-18T, DT-24, DT-24T, each end tested in series, eight series total, all at a shear span-to-effective-depth ratio of a/d=3a/d=3). All series except DT-24-2 failed in web-shear; DT-24-2 failed by anchorage bond failure at 840 kN versus 1400 kN for the web-shear failure of the companion end of the same specimen. Measured inclined-crack angles ranged from 26° to 47° (average about 35°), systematically shallower than the 45° truss assumption. Pretopped specimens were 6.5%–44.2% stronger than their untopped counterparts, with the precast flange and topping acting monotonically. Code comparisons showed most codes exceeding the test strengths for the 18 m specimens (ACI 318-14 about 1.5 times the test value for DT-18) while being 20%–30% conservative for the 24 m specimens, with Eurocode 2 the most conservative. The authors reported modeling the local effective prestress near the end region as increasing linearly from zero over a transfer length of about 3000 mm. Earlier work on double-tee shear (e.g. Ghosh's 1987 ACI study of shear reinforcement requirements for double tees) likewise did not address size effect.

The unresolved target is a mechanical/empirical model that predicts both the shear capacity and the governing failure mode of full-scale prestressed double tees, including the size effect, the end-region prestress profile, and the anchorage-bond failure mode that the source tests showed can govern.

Problem Statement

Establish and validate a predictive model for the shear strength of full-scale precast pretensioned prestressed double tees that resolves the discrepancies the source tests exposed. The model must (1) give the nominal shear strength with an explicit dependence on effective depth, so that the depth dependence is quantified rather than ignored, stating the sign of that dependence and the depth range over which it is supported and calibrating the term to the extent the 700 mm and 980 mm deep specimens permit (or reporting that those data do not resolve a decreasing-with-depth size effect), rather than treating a size effect as observed across those depths and extrapolable beyond them; (2) account for the end-region effective-prestress profile, making the assumed profile within the transfer length explicit and justified against available measurements, i.e. predict how the distance from the member end or support to the loading point relative to the transfer length, and hence the locally available prestress, affects the shear capacity; (3) reproduce the measured capacities and crack-inclination behavior of the Zhou et al. (2021) test series; and (4) provide a criterion demarcating the conditions under which anchorage bond failure between strands and concrete governs the capacity instead of web-shear failure — identifying the governing parameters (such as the prestress-to-concrete-strength ratio, available development length relative to the transfer length, and crack-width or slip thresholds at bond breakdown) — and correctly classify, for the end-region conditions of the DT-24-2 series, that anchorage bond failure rather than web-shear failure governs the capacity, without requiring numerical reproduction of the single observed 840 kN bond-failure load. The model must establish the parameter domain over which it is validated — including shear span-to-effective-depth ratio (all source tests used a/d=3a/d=3), amount of transverse reinforcement, concrete strength, prestressing level, and pretopped versus untopped flange configuration with the associated flange–web interface shear transfer — and identify the parts of that domain covered by full-scale data.

References

  1. Wei Zhou, Hao Li, Wenlong Zhang, 'Experimental behavior of the shear strength of full-scale precast prestressed double tees', Journal of Building Engineering 42 (2021) 102455. DOI: 10.1016/j.jobe.2021.102455. https://doi.org/10.1016/j.jobe.2021.102455
  2. Xueyu Xiong, Yang Zhang, Yipeng Ge, Hairong Shi, Moneef Mohamed Elobaid Musa, Gangfeng Yao, 'Shear behavior of full-scale prestressed concrete double tee beams with steel-wire meshes', Journal of Building Engineering 64 (2023) 105464. DOI: 10.1016/j.jobe.2022.105464. https://doi.org/10.1016/j.jobe.2022.105464
  3. Gangfeng Yao, Xueyu Xiong, Yipeng Ge, 'Cracking behavior of full-scale pre-tensioned prestressed concrete double-Tee members with steel-wire meshes', Journal of Building Engineering 44 (2021) 102658. DOI: 10.1016/j.jobe.2021.102658. https://doi.org/10.1016/j.jobe.2021.102658
  4. Hao Li, Wei Zhou, 'Flexural Behavior and Optimization of Full-Scale Prestressed Concrete Double-Tee Beams', The Structural Design of Tall and Special Buildings 35 (7) (2026). DOI: 10.1002/tal.70160. https://doi.org/10.1002/tal.70160
  5. Sun-Jin Han, Deuck Hang Lee, Kang Su Kim, 'Web-Shear Capacity of Prestressed Composite Inverted Multi-Tee Slabs', ACI Structural Journal 118 (3) (2021). DOI: 10.14359/51730538. https://doi.org/10.14359/51730538
  6. A. Chira, A. Puskás, D. V. Bompa, 'Load-deformation response of long span prestressed concrete wide beams', Journal of Building Engineering 65 (2023) 105631. DOI: 10.1016/j.jobe.2022.105631. https://doi.org/10.1016/j.jobe.2022.105631
  7. S. K. Ghosh, 'Shear Reinforcement Requirements for Precast Prestressed Double-Tee Members', ACI Structural Journal 84 (4) (1987). DOI: 10.14359/1655. https://doi.org/10.14359/1655