Subcontractors Disputes CA Transit Center Finds


On the heels of a report from San Francisco’s Transbay Joint Powers Authority that said multiple inspections failed to miss a construction error in the $2.2 billion Salesforce Transit Center, steel subcontractor Skanska USA is pushing back on that synopsis, saying it's premature to discuss the failure before the third-party peer review is complete.

What’s Been Happening

On Sept. 25, 2018, workers discovered a cracked steel beam in the third-level bus deck of the Transit Center, just six weeks after the structure opened to the public. The next day, another fissure was found on an adjacent beam.

The beams are part of the support system for the 5.4-acre rooftop garden and park, which includes a 5-foot layer of soil.

According to the Engineering News-Record, the bottom-flange cracks are near the 8-foot-deep midspan of each shop-welded girder. In the structure, the hanger plate slots through the bottom flange.

General contractors Webcor and Obayashi managed the center’s construction, which lasted from August 2010 to August 2018. The architect was Pelli Clarke Pelli, with Thornton Tomasetti serving as structural engineer.

In early October, the Transbay Joint Powers Authority confirmed that the shoring systems for the cracked sections were complete, making way for technicians to get in to take samples for analysis at LPI. Testing included scanning electronic microscopy, Charpy V toughness tests, Rockwell hardness tests, tensile tests, fractographic analysis and metallographic analysis.

A peer review was also initiated, overseen by the Metropolitan Transportation Commission.

In December, the report from LPI Inc., a metallurgical lab that investigated the issue, concluded that the problem started with weld access holes that were created in the beam during the joining process.

Also at that time, the transit center’s engineer-of-record, Tomasetti, presented a fix that calls for bolting 20-inch-wide steel cover plates above and below the area around each fractured bottom flange, similar to a “14-foot-long double splint.”

The independent peer review panel said that it was in “general concurrence” with the proposed fix, and on Jan. 10 the authority confirmed that it is in the process of gathering materials for the approved scheme.

The repair will be made not only to the girders with cracks, but to a pair of identical girders along First Street that have not fractured.

In February, it was reported that the steel plates were in the process of being fabricated offsite and were slated for a March delivery and installation. And, in a press release by the TJPA, all repairs have been scheduled and are expected to reach completion by the first week of June.

Late last month, TJPA executive director Mark Zabaneh revealed at a March 14 meeting that multiple inspections failed to miss a lapse in the building process. Had the error been caught, the steel beams never would’ve cracked, Zabaneh said.

Zabeneh reported that four layers of inspections—conducted by steel fabricator Herrick, steel installer Skanska, Webcor-Obayashi and the authority’s quality-assurance contractor Turner Construction—failed to detect a failure of a necessary part of the process.

Grinding the edges of the weld access holes smooth before welding can eliminate micro-cracking; Zabeneh said that had that failure been caught, the micro cracks never would have turned into the large fissures that compromised the integrity of the structure.

“The execution was not done properly, and that is something we are looking into,” Zabaneh said Thursday after an authority meeting. “It’s an area of great concern for us.”

What’s Happening Now

Skanska, along with Webcor and Herrick disagree with TJPA’s opinion that a construction error was the root-cause of the fissures.

Michael Pearce / Getty Images

Skanska, along with steel subcontractor Webcor Obayashi and steel fabricator Herrick Corp. disagree with TJPA’s opinion that a construction error was the root-cause of the fissures found within the hub’s two built-up plate girders above Fremont Street.

The main focus of the dispute is the 2-inch by 4-inch holes that were cut in the girders’ bottom flanges, located on both sides of the web, near the hanger plate.

According to LPI, the First Street girders didn’t fracture, unlike Fremont Street girders, because its holes had been cut after the welding process. Triggered by the report, new disputes over the definition of—and responsibility for—dimensioning and locating the design team’s term, “weld access holes” and the construction team’s term, “hole cuts.”

Stated in ENR, TJPA reports that it, along with Skanska and Herrick, was first introduced to the girder WAH by Tomasetti in a shop drawing review, where a “MISSING WAH” note was written on the drawing, which was then approved. The note also included an arrow pointing to the web in the area thickened by the hanger plate, just above the flange.

In LPI’s conclusion, had the WAH been installed to code, the fissures wouldn’t have happened.

However, Herrick President Bob Hazelton stated that these same holes that LPI are calling WAHs are not WAHs, but cuts in the flanges. His reasoning insists that WAHs are found in the web, not the flange, and were not required to access or complete a weld.

In addition, “these holes do not meet the size and shape of a weld access hole, which is why they were not ground or subjected to nondestructive testing,” said Hazleton.

As dictated in the American Institute of Steel Construction, the holes in question are not WAHs, as specified in its AISC’s Specification for Structural Steel Buildings.

With Skanska in fear that TJPA is turning a deaf ear to its pleas, it has held several meetings to express the concern over the girders’ unique configuration and high-stress design. But, a request to present these findings to the TJPA have since been denied.

In another attempt, Herrick wrote a letter to TJPA, dated March 20, registering its concerns about the terminology and methods used to evaluate the brittle fractures. TJPA replied, “[LPI CEO Robert] Vecchio accurately described the weld access hole appropriately.”

Skanska maintains that these details should wait to be discussed after the peer review is complete.

“The independent review will conclude what transpired, but whatever the conclusion, the public was let down and we intend to hold the responsible party accountable," Zabaneh said.

At the March meeting Zabaneh noted that the peer review of the tens of thousands of construction documents is reportedly about 85 percent complete, and no additional problems have been identified.

No announcement has been made of when the report will be released, but it is not expected any time before June or July.


Tagged categories: Accidents; Disputes; Maintenance + Renovation; Mass transit; NA; North America; Ongoing projects; Public Transit; Rehabilitation/Repair; Repair materials; Safety

Join the Conversation:

Sign in to our community to add your comments.