Skip to main content
Premier resource for practicing structural engineers

FRP Collector Strengthening in a California Hospital

Erik Moore, S.E., Aniket Borwankar, Sarah Outzen, P.E., Jay Love, S.E.
March 1, 2023

The Seton City Medical Center is located south of San Francisco in Daly City. The hospital, originally named Mary’s Help Hospital, was built in 1965 for the Daughters of Charity to support an underserved community in northern San Mateo County. The hospital remains dedicated to serving this community today but has fallen behind in meeting the California state-mandated seismic life safety performance requirements. As a result, the facility is under a tight 2023 deadline to meet the baseline seismic performance milestone set by the state to ensure their building can safely remain open for patients.

Background

The ten-story 209,000-square-foot hospital tower sits perched atop a hillside with commanding views of San Francisco Bay to the east and the Pacific Ocean to the west on the rare fog-free day. A segment of the North San Andreas Fault strikes past the facility less than a half mile to the west, capable of producing a mean characteristic 8.1-moment magnitude earthquake. The patient tower is rectangular in plan measuring approximately 200 feet by 75 feet, with concrete columns and 4-inch normal-weight concrete slabs that span one way to regularly spaced cast-in-place concrete pan joists. Lateral loads in the tower are primarily resisted by a central reinforced core wall that runs the length of the plan. The building is supported on spread footings founded on the sandstone rock outcropping that underlies the hill. The central tower core is embedded approximately fifteen feet below the basement level and supported on a continuous mat foundation.

In 2015, the hospital tower underwent a seismic evaluation as part of the California statewide mandate. The evaluation followed HAZUS, an approach originally developed by FEMA (Federal Emergency Management Agency) and adopted by the California Department of Health Care Access (HCAI, formerly OSHPD), which assigns a probabilistic risk of collapse to a building based on the site seismicity, building age, size, structural system, and specific deficiencies. HAZUS uses a checklist-based approach to rapidly identify deficiencies known to result in poor seismic performance. For example, one of the deficiencies identified was that three of the six tall, slender core walls have large slab openings for stairwells, elevators, or shafts on both sides. As a result, the building lacked adequate collectors to transfer lateral loads from the massive reinforced concrete floor system into these transverse walls. An initial seismic retrofit scheme addressed the collector deficiency with a design that used steel plates bolted below the floor, to the sides of concrete pan joists. The proposed system was designed to collect load from the diaphragm and drag it back to the transverse walls with high-strength all-threaded rods cored through the longitudinal walls.

HCAI issued a permit for construction in mid-2016 to retrofit the building and address all the deficiencies identified in the HAZUS evaluation. The retrofit included these collectors, strengthening to address non-ductile concrete columns, discontinuous concrete shear walls, brittle precast wall panel connections at the podium, and foundation strengthening to stiffen the building against lateral loads transverse to narrow tower core. Unfortunately, the facility was forced to put the seismic strengthening project on hold after suffering financial setbacks in 2018 and a subsequent change in ownership.

Project Restart

In the spring of 2020, the seismic strengthening project restarted under an even tighter schedule after new owners purchased the facility. The facility must complete construction of the seismic strengthening by the state-mandated deadline of 2023 or risk fines, loss of their operating license, and possibly forced closure. The owner tasked the contractor and design team with reviewing the permitted retrofit design to identify opportunities to reduce the schedule.

The team identified the steel plate collectors as one point of construction schedule risk. Although the team conducted pre-construction surveys to verify field conditions for the design, it was well understood that the fabrication of the steel plates and installation would require extensive coordination in construction. The alignment of pan joists in the floor system relative to the transverse walls varied up the height of the building resulting in unique alignments of steel joist plates, brackets, and threaded anchor rods; each would need to be surveyed, verified, and fabricated. Installing the plates in the cramped overhead ceiling would also pose a challenge to drill anchors and hoist steel plates into place, not to mention working around the extensive utilities crowding the ceiling in these locations. It would require a series of complicated shutdowns and rerouting of existing utilities to provide the access needed to install these steel plate collectors.

The design team identified externally bonded fiber-reinforced polymer (FRP) collectors as a viable alternative to overcome these challenges and keep the project on schedule. The team reached out to two vendors to provide designs for the collectors and concrete column-wrapping using FRP. Simpson Strong-Tie was selected based on their willingness to perform project-specific installation mock-ups, and testing at their research lab in Stockton, California. Although carbon fiber has been used for over 30 years in column retrofits, the use of the material for diaphragm and collector strengthening is still governed by reference documents and has not officially been adopted by the California Building Code. This is primarily due to the lack of any substantial testing for multi-layer FRP ties, similar to highly loaded collectors. Shortly after the Seton collector redesign began, Simpson Strong-Tie published its first International Code Council (ICC) approval for FRP materials in diaphragms and collectors (ESR-3403, 2021). To use FRP collectors on this previously permitted project, the team requested an Alternate Method of Compliance from the Seismic Compliance Unit at HCAI. After discussions with the Seismic Compliance Unit, Simpson Strong-Tie volunteered to perform physical testing at their lab to prove the use of their materials for this application.

Testing and Results

Current design standards for externally bonded FRP provide limited and sometimes differing guidance on the design of collectors. Debonding strain (from the concrete substrate) is a key limit state in the design of FRP collectors and can vary with both the number of FRP layers used and the presence of FRP anchors to the slab. ICC-ES AC125, Acceptance Criteria for Concrete and Reinforced and Unreinforced Masonry Strengthening Using Externally Bonded Fiber-reinforced Polymer (FRP) Composite Systems, provides guidance on un-anchored FRP systems with an equation for debonding strain as a function of the compressive strength of the concrete, number of FRP layers, thickness per layer, and elastic modulus of FRP reinforcement. This equation is identical to the one presented for flexural strengthening in ACI 440R-17, Externally Bonded FRP Systems for Strengthening Concrete Structures. The International Association of Plumbing & Mechanical Officials’ IAPMO EC 038, Evaluation Criteria for Diaphragm Strengthening Using Fiber-Reinforced Polymers provides guidance on anchored FRP using a different equation to determine the debonding strain limit, identical to the equation presented in ACI 440R-17 for shear strengthening applications. Although EC 038 refers to “fully-anchored FRP,” it does not clearly specify these anchoring requirements. Simpson Strong-Tie proposed a test program to address differences in these design standards and to better determine criteria specific to the highly loaded, multi-layered FRP collectors needed for this project.

Simpson Strong-Tie conducted scaled experimental testing at their Tyrell Gilb Research Laboratory, an ICC-ES and IAPMO-accredited laboratory. The testing program consisted of 24 reinforced concrete and FRP collector specimens with differing numbers of layers of FRP fabric, anchor diameters and spacings, and concrete substrate strengths. A hydraulic actuator applied direct tensile monotonic loading to the free end of the FRP fabric. Simpson collected actuator load and displacement data during each test to calculate composite FRP collector stress and strain response. Simpson also used a non-contact 3D digital image correlation (DIC) system to capture a pair of digital photos every second for full-field displacement and strain measurements. A research paper detailing the experimental testing was presented at the 2021 SEAOC (Structural Engineers Association of California) convention (Hosseini et al., 2021).

The test program determined that existing design equations overestimated the FRP debonding strain for multilayer unanchored collectors (the design equations have since been revised based on this and other testing). The results also confirmed that regularly spaced FRP anchors significantly improved the performance of the collectors by allowing load sharing between multiple anchors along the length before the fabric debonded. Additionally, larger diameter anchors increased the capacity of multilayer FRP collectors, with anchor shear rupture becoming the governing failure mode for these highly loaded specimens. After review and discussion among Degenkolb, Simpson Strong-Tie, and HCAI, the team established project-specific design criteria. FRP anchors were required for the full length due to the poor performance of the unanchored collectors observed during testing. Rather than using the design equations in AC 125 or EC 038, strain limits were derived directly from test results. At highly loaded ends of the collector (near shear walls), anchors were required at 16 inches on-center with a maximum debonding strain limit of 0.0028in/in. At the far ends, anchors could be spaced up to 3 feet on-center with a lower strain limit of 0.0015in/in.

Installation

Simpson Strong-Tie continued to provide support through construction for the FRP collectors as the delegated design engineer of record, working closely with the design team and contractor. With each challenge encountered in construction, the flexibility of the FRP material proved valuable for schedule savings. Rather than extensive redesigns and needing to refabricate steel plates, the FRP collectors could be adjusted to pass around obstructions like fire risers with additional patches of fabric and FRP anchors. When the ends of the transverse concrete core walls were too congested with existing rebar to fully embed the large diameter FRP end anchors, the team could provide alternate designs, with additional FRP fabric passed through the longitudinal wall and anchored to the face of the transverse core walls.

This flexibility allowed the collectors to keep pace with the rest of the construction schedule. As the team approached the tower floors, the installation fell into a regular pace. Timelines included approximately 7 days to put up infection control and demolish the existing flooring and the base of partition walls, 6 days to install the actual FRP fabric as well as the cementitious FX-207 fire retardant coating, and 17 days to complete the build back for flooring, partitions, and final punch list. While walking the construction site to view the collector progress, a clear contrast could readily be seen between the relatively unobstructed installation of FRP fabric on the top of the slab versus the tangle of MEP systems above the ceiling that the team would have had to navigate to install the originally designed steel plate collectors.

Conclusion

The final FRP collector was installed in October 2022. Although there were challenges with additional testing and approval of the FRP collector design, the savings in schedule and construction risk associated with working around the existing utilities in the ceiling made this a clear decision. Implementing this innovative solution for a California hospital was made possible by the openness and collaboration of the whole project team and the Seismic Compliance Unit of HCAI.

Project Team

Structural Engineer: Degenkolb Engineers, Oakland, CA

FRP Engineer: Simpson Strong-Tie, Pleasanton, CA

Architect: Smith Group, San Francisco, CA

General Contractor: Swinerton, San Francisco, CA