Skip to comments.Bay Bridge inspections: busted bolts
Posted on 03/27/2013 9:25:11 AM PDT by NormsRevenge
At least 30 of the giant bolts that hold together the new, $6.4 billion eastern span of the Bay Bridge have snapped.
As a result, Caltrans is considering replacing all 288 of the bolts on the new bridge before it opens, The Chronicle has learned.
Caltrans insists the new span is safe and that plans to open it the day after Labor Day are still on track.
However, officials say it's too early to determine how long it will take to fix the problem - or the cost.
Toll Bridge Program Manager Tony Anziano said engineers are "pretty confident" the problem with the bolts is not a design issue or a construction problem but related to the quality of the steel bolts themselves.
"This isn't exotic - this isn't some wild issue," Anziano said.
Unlike the Chinese-built deck sections, the bolts - some as long 17 as feet - were produced in the United States.
"It appears to be a type of materials problem - the presence of hydrogen in the metal," he said. The hydrogen makes the metal brittle.
(Excerpt) Read more at sfgate.com ...
This bridge is supposed to take an 8.5 or such.. we'll see soon enough if the Ring of Fire pops one off here..
These seventeen foot long bolts were NOT tested prior to installation?
What’s that smell I smell?
I once met a guy who sold nuts and bolts.
We discussed them for maybe a couple of hours. The markings, quality, the ones used for extreme duty etc. Far more interesting than one might think.
Unfortunately that was over 40 years ago and I don’t remember much. I do remember he said there are a lot of injuries caused by people replacing the nut which holds lawn mower blades on. He said they sometimes just use mild steel nuts or bolts or whatever they are.
the bolts - some as long 17 as feet - were produced in the United States.
"It appears to be a type of materials problem - the presence of hydrogen in the metal," he said. The hydrogen makes the metal brittle.
I'm curious what is the origin of steel that went into those bolts? Does the "produced in the US" tag only mean that they were machined here?
Recall that government presumes that it is competent to command a monopoly on bridge certification and inspection. If a private company had this gross failure of inspection and maintenance, the very same government officials would be rushing in front of the cameras to denounce this laxity and to tell the public how government is protecting them.
Doesn't seem very concerned, does he? Why can't they do their QC and test the parts before installation. Disclaimer- I am no engineer.
Ban fat lesbian crossings.
Pretty confident. Those are words I don't want to hear about any building, bridge, or aircraft.
It seems like everything made today is crap and things made a long time ago are light years ahead.
I believe Kiewit (used to be Peter Kiewit & Sons) is the builder. They are also the ones making the pontoons for the new Lake Washington floating bridge, which are a mess. Leaking, rebar rusting, rebar mislocated or missing completely, and the state is making excuses for them and wasting our money, not to mention the safety of those who will be driving on this bridge.
That information alone made FR worth reading today.
Sounds like a possible engineering problem and the amount of torque used when installing the bolts.
The S.A.E. has ratings on bolts, and it seems the engineering for the bridge should have allowed for worst case conditions -- not best case conditions.
In the 1970's, I worked for a computer company (Microdata) that built a computer based on best case conditions for circuits. It did not account for slow parts.
Of course, semiconductor devices were just started to be used for memories as well as other devices.
The original 1K MOS memories were terrible devices that came out of Silicon Valley then...
You *never* buy 200+ 18' bolts without specifying their performance. Each bolt probably costs more than $10K, so it's a multi-million contract.
Here is the machine that you use for testing metals for strength. This is just a well instrumented hydraulic press, backward. Samples of special shape are made, and then the machine pulls on them, accurately recording the tension and the deformation. This gives you the stressstrain curve. Eventually the sample breaks.
Even if the company that made bolts doesn't have such a machine, it would be sheer insanity to buy a hundred tons of steel without sending a sample to an independent lab for these tests. You would want a chemical and a crystallographic analysis in addition to that. This is what the engineers did after the bolts failed; that's why they are telling us that the steel was wrong (not the right type, or the right type made incorrectly.)
Now the manufacturer of those bolts is going to lose their shirt on this deal. I'm sure their profit margin is not fat enough to cover redoing all bolts and still staying in black.
“I’m pretty sure this gun isn’t loaded.”
“I think your parachute is packed right.”
“I’m pretty sure I got all the sponges out of your body cavity.”
“Almost positive there’s enough cash to cover the check.”
Where have all the adults gone? Is there really a Galt’s Gulch?
That I didn’t get an invite means I better step up my own game.
They came loose because they were elongating beyond elastic limits or even breaking under tension.
Defective parts or defective design? More stress on the bolts than the engineer predicted?
Apparently, they had concrete deck sections MADE IN CHINA that has caused quite a bit of trouble in the 10 year history of this bridge.
Cheap things from China is also a serious problem in this bridge: concrete sections fabricated in China caused considerable delay in the 10 year history of attempting to get the bridge built.
In fact, according to the article, it sounds like trying to save money having the concrete deck sections MADE IN CHINA has probably cost more money in the long run than doing it right in the United States.
This is not even the stress of an earthquake causing these problems, and as someone mentioned, the bridge is supposed to withstand an 8.5 magnitude quake...
>> Sounds like a possible engineering problem and the amount of torque used when installing the bolts.
Or unexpected forces on the assembly.
How does one accurately torque an 18’ bolt?
He is right in a sense- you build and you test.
Usually you end up confirming your design is OK, but occasionally you find an issue like this.
That is why you test.
This poster claims it is sheer insanity to spend millions of dollars on bolts and not QC them.
Bolts made in China?
Or thermal expansion. The longer a bolt is, the more it grows and shrinks in response to heating and cooling, and these were 17'.
Sounds like in the former Soviet Union....
The nail factory made their quotas every year....
increasing the size of the nails....(big ones are easier to make)
Ended up with a huge pile of 50cm long nails that nobody brought.
It is standard to bake the bolts at about 600f after plating to remove the hydrogen
May have been hard if they did not have a 18 foot long oven
Can be both. However design of bolt- or rivet-connected bridge spans is a very well known process. Bridges of this type were built for more than a hundred years, entirely without computers; all you need is printed tables. Steel made in China, on the other hand, with no quality control and no incoming inspection, is a very likely culprit. An engineer doesn't claim that "the steel has too much hydrogen" until he has the lab test results in hand.
A comment above mentions that some bolts were loose. If the holes are too large this can result in uneven loading of bolts - and then indeed some will be sheared off, and then the rest follows. If the bolts are not tight then they will experience axial stress instead of shear stress, and the thread will be torn off. It's all very obvious when you look at the bolt, and the guys who are investigating are usually experienced troubleshooters.
Caltrans insists the new span is safe...
I still distantly miss the beaches once in a great, great while.
About seven years ago, the oil field went through a period of buying Chinese drill pipe. Failure after failure after failure.
Not that hard to contract out. Friend of mine works in a place with an industrial oven several hundred feet long.
When you torque a bolt, even the toughest alloy, it stretches almost like a rubber band. It has to hold that state of tension without elongating beyond a certain point and loosening, or breaking, and it must hold this property over time. Design involves choosing the proper material for the expected load but sometimes the best calculations fail in practice.
The problem in California is not the bolts, it’s the nuts.
Considering what happened with the Nimitz Freeway pancaking during an earthquake, going above-and-beyond should be done in a place like San Francisco.
The reference to the pancaking of the Freeway was that the freeway had two layers. The earthquake caused the top layer to come down on the bottom layer. Some folks were crushed in their cars as a result, as I recall.
They will all fail, given time. You can extend the time but it will affect other considerations, appearance, cost, etc.
You could, and should, do a simple nondestructive hardness test on the material being used to make the bolts before the bolts are made and before any other test are done.
You could do a hardness test on enough material to make 288+ bolts in about 2 days max.
That would tell you if the material is up to specs before wasting any more time or money on it.
Too soft or too hard chunk it.
No point in spending any more time or money making a bolt when it’s going to be junk anyway.
You have to understand that this is a Cal Trans project, and Cal Trans is one of the most incompetent organizations within the State of California.
Several years ago in Orange County on a major freeway interchange where Cal Trans was in charge of constructing freeway “flyover” ramps, they discovered at a very late stage that the pre-stressed concrete was below acceptable quality. Initially they thought that the entire set of flyover ramps would have to be torn down. They finally found a way to “rework” them but it took nearly two years to do so.
In the case of this bridge, there was no excuse for both the general contractor and Cal Trans to have strength tests run on those bolts as well as all major steel components. But what the hay, this is California, and that just about says it all these days.
In the 70’s and 80’s it was American steel that was junk.
In one job alone I rejected over 1 million feet of 2 3/8” tubing.
When I’m “pretty sure”, and I often am, I check.
The GG Bridge was built in the early 1930's. In the 1980's, for the 50th year celebration, they allowed a mass of people to walk the bridge, completely filling it with people (jammed together weighing much more than any load of vehicular traffic). The bridge flattened out from it's arch but everything held together, and engineers were able to relax from their fear of a collapse (politicians wanted the people walk). That's 1930's U.S. steel! Nowadays, replacement steel pieces regularly break on the Oakland side of the Bay Bridge, some having damaged cars. Part of the reason for the new section.
I was in that business for about a decade. Structural grade bolts are subject to stringent requirements. In fact about 20 years ago cheap Chinese counterfeit bolts were the cause of a collapse in a shopping mall in St Louis IIRC.
Selling counterfeit or mismarked fasteners is now a Federal crime.
Thank you most kindly for your informative response.
You must be a youngster.
All I remember is 12,000 mile spark plugs, 20,000 mile brakes, and 20,000 mile tires.
A 100,000 mile car was a real outlier. Now that's the average car on the road.
I am an electroplater and metal finishing engineer.
I do not know of any plater in the USA with a 17’ oven. If the bolts are 147KSI in tensile or harder than RC32, they should have been baked for at least 4 hours before plating and a number of hours after plating. The post-plating bake is determined by hardness or tensile. I would have adjusted the bake for the thickness of the part since you have to bake the hydrogen out of the center.
The plater could have sent the parts out to a heat treater with a larger oven, but this has to be done quickly after plating before the hydrogen in the part has a chance to embrittle/damage the steel.
It is possible that the steel could have been embrittled by hydrogen before the parts got to the plater. Certain cutting oils can release hydrogen or it could have been pickled/de-scaled using acid which introduces hydrogen. However, the most common source of hydrogen is corrosion. Letting the raw stock rust is dangerous. After the hydrogen has done its embrittlement dirty work, the stock is ruined. You cannot fix it short of resmelting, reforging, re-heat treating, etc.
I have seen a lot of poor quality steel come through, some of it from U.S. mills, most from China.
As for testing, NASA and the military contractors have to do it, so why not the bridge builders?
This entire incident shows the need to have quality inspectors at every supplier down the line when you are dealing with life-critical parts.
When we built the Space Shuttles, we sent inspectors to the MINES to see where the ore came from and to take samples for the lab.
Let me see here...
“At least 30...”
“all 288 of the bolts on the new bridge”
Comes to somewhat over 10%, doesn’t it?
“Caltrans insists the new span is safe...”
Suuuuure it is....
Seismic designs today are premised upon strength in ductility. Brittle failures were common in the Northridge quake, which changed many structural codes and design philosophies.
It might be something more fundamental like Min Gauge length or the distance the strain must be transmitted to the bolt for it to behave as a connector.
I wonder if the American factory hired illegal workers.
In any event, between these bolts and the Chinese-made parts, when the Big One hits, this sucker’s going into the water.
Better engineering and quality assurance management would have avoided this.
Good to catch it now, however.
The bolt weren’t specified well enough, in terms of strengths and/or tests were not made prior to installing said deficient parts.
Somewhere in the bureaucracy perhaps too much diversity trumped engineering & construction practices we used to handle fine.
Maybe the bolts came from a government ordered procurement from a minority owned supplier.