The engineer who designed the troubled fix for the tilting and sinking Millennium Tower has told city officials that a test planned for this month could result in a quarter of an inch of new sinking -- a risk one outside expert says is simply too big given the apparently shrinking safety margin for the building.
Last month, crews bored holes at the northwest corner of the building, where the luxury high-rise is sinking and tilting the most.
The test was designed to sample the soil and to install monitors to measure the extent of vibration from installation of new piles to support the building on two sides. That installation was halted in August after the building began to suddenly sink and tilt more.
New data shows that while the extent of settlement subsided during that work stoppage, it accelerated during the week of September 20, during that new test boring work. In just that one monitoring period, the tower sank one 25th of an inch –which translates into a rate of two inches a year.
Get a weekly recap of the latest San Francisco Bay Area housing news. Sign up for NBC Bay Area’s Housing Deconstructed newsletter.
That rate has since leveled off. Engineer Ron Hamburger, who designed the so-called fix, is now calling for a far more elaborate test -- sinking a new support pile to bedrock, although he concedes there’s a risk of more settlement.
Given the uncertainties, one expert says that’s too big a risk.
“We’re missing something here, right?” said Rune Storesund, executive director of UC Berkeley’s Center for Catastrophic Risk Management. “So we need to stop, figure out what it is we’re missing. Otherwise, we’re just kind of guessing what’s going to happen in the future.”
Monitoring data reflects an excess of one inch of settlement since work began on the fix project in May. So far, crews have dug holes for 33 out of the originally planned 52 piles.
As a result, the building is now tilting 5.5 inches more to the west and Fremont Street and 2 inches more, north toward Mission Street.
In his recent letter to the city seeking permission for the pile installation test, Hamburger establishes boundaries for how much tilting could occur while maintaining the buildings “safety and stability” in an earthquake.
Currently, the tower is tilting 22.5 inches towards Fremont and 9 inches toward Mission. Under Hamburger’s analysis, the building can tilt 6.5 inches more to reach the 29-inch threshold on Fremont and three inches to reach the 12-inch Mission threshold, and still remain seismically “safe and stable.”
Beyond those levels, he suggests, the building’s performance in a quake will start to suffer.
Hamburger predicted that the test pile installation work could trigger as much as a quarter of an inch of new settlement – resulting in about an inch of new tilting at the top toward the west and about a quarter of an inch to the north.
“We do not anticipate that any settlements likely to occur during the pilot installation will pose a threat to the building,” Hamburger assured city officials.
But UC Berkeley’s Storesund worries that one inch of additional settlement does not provide much of a margin of safety.
“If we only have an inch left before things take a turn for the worse, there’s not a lot of room for error,” Storesund said.
“This close to a boundary between safe and not safe – yeah, that should be really, really worrisome,” said Keith Porter, an engineer and adjunct professor at the University of Colorado, Boulder.
Porter, who has studied and developed seismic performance standards for buildings, says the “safe and stable” margin as defined by Hamburger is so narrow that fix designers should be planning for what to do should the tower tilt beyond that threshold.
“Are we just hoping? Are we going to do this and hope nothing bad happens – and if something bad happens, say ‘oh well,’” Porter said.
For now, Hamburger is assuring city officials that an engineer will have the authority to stop the test work “if the predicted settlement effects are exceeded.’’