The California Department of Transportation said Monday that it was reviewing whether the large bolts that broke last month on the San Francisco-Oakland Bay Bridge had failed quality-control tests by the agency's materials lab.

Those bolts, manufactured by Dyson Corp. of Ohio, were used to secure shear keys and bearings, vital seismic safety equipment on the eastern pier of the new suspension span.

Dyson Corp. built at least 14 other parts for the suspension span job that have failed Caltrans quality tests - far more failures than any other supplier on the contract, according to Caltrans records obtained by The Bee. Normally, parts that fail such tests are rejected.

Asked whether the broken bolts failed quality tests, Caltrans spokesman Will Shuck said Monday, "that's all under review at this time."

The prime contractor is a joint venture of American Bridge Co. and Fluor Corp. Caltrans "expects contractors to select suppliers and subcontractors who will get the job right," Shuck said. "The department has ordered a full review to find out exactly what went wrong, to determine how it will be fixed and assure that responsible parties are held accountable."

Shuck did not say whether Dyson's performance record had been a concern to Caltrans.

According to Caltrans records, the parts manufactured by Dyson that failed quality tests included 1-inch diameter, 17-inch long bolts - also for the shear keys and bearings - and 3.5-inch diameter, 4-foot long anchor rods for the main cable of the suspension span.

Caltrans released a memo written by its director, Malcolm Dougherty, directing Toll Bridge Program Manager Tony Anziano to "identify any other components of the bridge that came from the same manufacturer or supplier and re-inspect."

Brian Rawson, Dyson's chief executive, could not be reached for comment.

The Bee obtained records of the tests of 352 parts for the suspension portion of the bridge. These involve 61 subcontractors to American Bridge and Fluor. While not comprehensive - for example, the broken bolts reported last month are not part of the record - the information shows that Dyson faced consistent quality control problems.

Nearly half of lab test failures involved parts provided by Dyson exclusively, or were part of a job completed jointly by Dyson with another subcontractor. No other company identified in the records had comparable test failures in number or rate.

Professor Arthur Hucklebridge, an expert on seismic issues in bridge construction at Case Western Reserve University in Ohio, said using parts that had failed quality testing would be unusual. If that has been done, he said, "they probably had to have some pretty good justification - for example, they didn't accept the necessity of that particular test, or thought that it wouldn't impair the function" of the bolts.

On Monday, The Bee requested a complete set of test documents for the broken bolts. The agency said it could not respond to the request because of the state Cesar Chavez holiday, but would do so at a later date.

State Sen. Mark DeSaulnier, D-Concord, chairman of Transportation and Housing Committee, said that unless there is a legal or personnel issue that prevents disclosure, the agency should release the documents promptly.

Dougherty, directed Anziano to conduct a comprehensive review of what went wrong with the bolts, including a request to "articulate the testing and compliance certification process employed during construction to fulfill Quality Control/Quality Assurance efforts."

The memo also asks Anziano for "a complete forensic analysis" and to "put the contractor on notice in writing that they need to demonstrate that they are in compliance with contractual requirements and secure their proposed course of action."

Officials said last week that hydrogen contamination, either during production or during the process of coating the steel bolts with zinc to galvanize them, could have made the parts brittle. Caltrans has not yet stated if Dyson or some other party is responsible.

The 288 suspect bolts, also called anchor rods, are 3 inches in diameter and up to 24 feet long. They attach the four shear keys and four bearings to the eastern pier of the suspension span. Among those, 32 bolts snapped after they were tightened.

Dougherty's memo indicated that the broken bolts were installed in 2008, while some others were installed in 2010.

Caltrans officials remain hopeful that the new $6.4 billion eastern span will open on Sept. 3, but Anziano acknowledged last week that the broken-bolt setback could delay that schedule.

Dougherty reinforced that point in his memo.

"I want to reiterate that safety of the bridge once opened is the paramount and controlling factor in all decision making on this issue," Dougherty wrote. "We will ascertain any cost and schedule implications once the best solution consistent with the design life of the bridge and performance expectations are met."

Last week the agency said it would consider a collar-like addition to at least two of the shear keys, which help control lateral movement of the bridge during a major quake.

That solution, if adopted, would take months to design, build and test.

Call The Bee's Charles Piller, (916) 321-1113. Follow him on Twitter @cpiller.

© Copyright The Sacramento Bee. All rights reserved.

• Read more articles by Charles Piller

Order Reprint