Delays plague high-speed rail bridge near Madera. What’s gone wrong in construction?
For some residents in the Madera Acres neighborhood north and east of the BNSF Railway freight tracks, what would ordinarily be a rather straightforward 3 1/2-mile daily trip down Road 27 / Lake Street into downtown Madera for work or school has, for much of the past several years, required a detour that for a time stretched to almost 11 miles.
Resourceful neighbors have since become familiar with a shortcut that pares the drive to less than seven miles – still about double what it was before Road 27 was closed for construction of a high-speed rail road overpass that has had its share of troubling problems and errors by the joint-venture contractor, Tutor Perini/Zachry/Parsons.
“They had a meeting and told us it was supposed to be done at the beginning of this year,” said Jose Gonzalez, a 19-year-old who has to make the extended daily trip into Madera for work. The Gonzalez family lives on Tremaine Avenue, just northwest of the bridge construction.
That once-anticipated opening date came and went; Gonzalez said residents have not received an update from the California High-Speed Rail Authority on a target date for completion and reopening the road. The agency’s chief executive said a new estimated completion date is about a year away, in summer 2021.
Work at the site has been hampered after steel wire “tendons,” intended to provide structural strength to the overpass, began to snap and fail without warning in October 2019. In more than 10 months since, the California High Speed Rail Authority, the contractor and consultants whose job is to inspect and oversee the quality of the construction work have all grappled with why the cables failed and how best to fix them.
The rail authority’s board of directors will conduct a hearing on the bridge concerns when it meets Wednesday in a teleconference, about three weeks after authority CEO Brian Kelly and two agency board members from Fresno – acting chairman Tom Richards and Henry R. Perea – visited the site Aug. 17 for a firsthand look.
“It’s an unacceptable thing that needs to get remedied and needs to be remedied as soon as we can,” Kelly told The Bee in an interview after his visit to the bridge. “It’s unfortunate, it’s unacceptable from an owner perspective. We’re going to work with the contractor to make sure those strands get replaced.”
The current state of the 637-foot overpass – which is about 300 feet or so from the Gonzalezes’ back yard – suggests that there’s quite a ways to go before the road can reopen. The bridge span is in place atop several sets of massive columns. But the ramps that connect the bridge span to the closed roadway have yet to be built.
It’s not the ramps, however, but what lurks within the structure of the bridge that has ground much of the substantial work to a halt.
What went wrong?
Supporting the road deck that will at some point carry traffic above the railroad tracks, four concrete girders span the 637-foot length of the bridge, about two football fields long. Five hollow tubes run the entire length of each girder. Each of those tubes, in turn, is supposed to encase a “tendon” of 15 steel strands more than a half-inch in diameter. Each strand consists of seven individual wires. When anchored and tightened with tension, the 20 tendons – a total of 300 strands – provide structural strength to the bridge.
Documents from the rail authority indicate that one requirement of construction was for the tubes or ducts to be filled with grouting material within days of installing the steel tendons to seal out moisture and adding further strength to the tendons.
The first sets of tendons were installed in the girders in April 2019 – but the ducts were never grouted. About six months later, in October, some of the strands began to snap and fail in two of the girders. Two two weeks later, the state rail agency issued a “stop work” order for the contractor’s work on the tendons.
An engineering investigation commissioned by the contractor and prepared by Illinois-based Wiss, Janney, Elstner Associates Inc. concluded in March 2020 that water made its way into the tubes and stayed there, leading to corrosion of the steel strands and ultimately compromising the tendons.
The investigation report includes photos taken with a camera pulled through the tubes after the failed tendons were removed showed sitting water along the bottom of some tubes, as well as evidence in one tube that some steel tendons had been partially submerged – and in one tube, almost fully submerged – for at least some period of time.
Why did it happen?
Emails among engineers within the rail authority indicate their belief that the cause of the tendons’ failure lies in something else that happened – or, rather, what didn’t happen – between April 2019, when the tendons were installed, and October 2019, when they began to fail.
At some point in construction, vent caps were damaged and some of the vent pipes leading into the tendon tubes “were damaged or sheared off, and this would likely allow water intrusion into the ducts,” states a March 19, 2020 email by Mark Barresi, a quality oversight manager for PGH Wong Inc., part of Wong + Harris, a joint venture hired to perform construction project management oversight on behalf of the state.
But another March 19 email by David Vallejos, a project construction engineer for the rail authority in Fresno, suggests the problem was a failure by the contractor to meet construction requirements.
“TPZP states that the root cause was water entering the ducts,” Vallejos wrote in his note to PGH Wong members including Barresi and others. “In my opinion, the root cause is that TPZP didn’t follow the plans and specifications by not grouting the ducts as required.”
Kelly told The Bee that he saw two major problems “that really trouble me” with the Road 27 overpass: First, that the vent caps were damaged “in a fairly careless way,” allowing water to enter the tendon tubes, and second, improper storage of the strands on the construction site before they were inspected, in containers that got wet during a rainstorm.
“This is, to me, indicative of quality on the site that needs to get better,” he said. “Part of our response to this is not just, ‘What went wrong at Road 27, we need a fix,’ but also, ‘What’s going on with quality that we need to improve?’”
In the emails that have flowed between the contractor, project construction managers and officials with the rail authority, what’s not explained is how months went by without the ducts carrying the tendon strands being grouted before the strands began to break – despite several layers of quality-control oversight.
Tutor Perini/Zachry/Parsons is the contractor hired to build Construction Package 1, a 31-mile stretch of the rail route from north of Madera to south of Fresno. When the contract was approved in 2013, it was valued at just under $1 billion. Delays and change orders in the intervening seven years have driven the price to more than double, at $2.2 billion, of which about $1.2 billion has been spent as of July.
The authority also has a $112.8 million contract with Wong + Harris to provide construction management services for Tutor Perini’s work on Construction Package 1.
Beyond that, the authority is paying WSP USA Inc. for overarching services as its “rail delivery partner” to steer work on all three construction contracts now underway in Madera, Fresno, Kings, Tulare and Kern counties and planning for future work throughout the state. The WSP contract is valued at more than $666 million.
Kelly dismissed suggestions that the damage to vent caps during construction, failure to grout or seal the tendon ducts and water corroding the strands represent a systemic breakdown of oversight or should have been caught earlier.
“This got caught. We started to respond to this a week after it occurred. It was the authority itself that, with our oversight and (Wong + Harris), issued the work stoppage,” he said. “Our quality management team is making sure the remedy is correct.”
But an authority audit of Tutor Perini’s quality oversight team showed that “they did not do well enough,” Kelly added. A new “corrective action plan” calls for the rail agency to bring in engineers and inspectors from the California Department of Transportation “to ensure inspections are done more thoroughly and to ensure that these kinds of things don’t happen,” he said.
What’s the solution?
Tutor Perini had already replaced the tendons that failed in two girders, but as with the first sets, those also were not grouted.
Replacing the failed strands is not going to be enough, Kelly said. This past April, the rail authority issued a requirement that Tutor Perini replace all 15 strands in all of the tendons in each of the four girders, and not just the tendons that failed.
“All the strands that are in there now come out. Everything gets completely dried out, …,” he said. “We have to confirm that the tubes are all in good shape, which we will do with cameras. Then we will secure the vent caps at the top, put in new strands and immediately start grouting.”
Kelly said he expects Tutor Perini to bear the costs of the repairs rather than adding it to the already increased construction contract.
“It was an issue in construction; that’s on them,” Kelly said. “The issue with water getting in there comes back to them. Certainly our position is this needs to be repaired, and it needs to be repaired on their dime.”
Kelly added that he believes the repairs can be accomplished sometime this month and allow additional construction to move forward. “This work doesn’t actually take that long; we just have to get the steps done,” he said. “Then we have construction on both ends that we have to finish.”
For residents who have waited for years to return to their regular route to and from Madera on Road 27, Kelly said, the new target date for completion and reopening the road is summer 2021 – five years after work began. But, he added, “that bridge will not open until we are certain that it’s safe. That’s the threshold for us.”
What about other bridges?
While Kelly said he believes the oversight process ultimately worked on the Road 27 process, in the days following the initial failure of the steel tendons at least one of the authority’s other top executives expressed concern about other work done by Tutor Perini/Zachry/Parsons along its 31-mile construction segment in Madera and Fresno counties.
In late October 2019, before Wiss, Janney, Elstner was asked to investigate the strand failures, authority Chief Operating Officer Joe Hedges indicated in an email that he wanted to know what caused the strands to fail before allowing Tutor Perini to move forward with its planned repairs.
“Repair of broken strands is just one piece of (noncompliance),” Hedges said in an Oct. 28 email to the authority’s own engineering team, project construction managers at Wong + Harris and WSP, and TPZP executives including Tutor Perini CEO Ron Tutor. “The piece … that I am most interested in and most critical element (is) ROOT CAUSE OF FAILURE.”
“Currently, every structure in (Construction Package 1) is in question until root cause of failure is determined,” he added. “Losing six strands in a single beam makes little engineering sense.”
Public records released by the rail authority indicate that several prior “noncompliance reports” preceded the problems with the tension strands over the past few years, from encroaching on BNSF Railway right of way for construction access to the site and minor alignment issues for the concrete support columns, both in 2017; to a 2018 incident in which a crane operator lifting one of the concrete girders into place struck another piece of the bridge structure.
In mid-2018, the authority acknowledged that it issued a stop work order for another bridge that was being built by Tutor Perini at Avenue 8, southeast of Madera and required the builder to tear down retaining walls that had already been built and rebuild it with a different design.
This story was originally published September 8, 2020 at 5:00 AM.
