Electric railways versus megathrust earthquakes

[erikem]

PDN: This has been a long time question for me. Since I know nothing about the design of bridges I only have observations.

The Loma Prieta quake broke several concrete bridges notably the I-880 collaspe but also some rather minor concrete bridges. At least one near Aptos (epicenter). Even though the Oakland Bay bridge did not have a design quotient for the magnitude of the quake only one top steel bridge section fell onto the lower section under it with rest of bridge left standing.

The Northridge quake had many concrete highway bridges collaspe but the SP line from LAX to Palmdale was up and running in just a few days. (has several steel bridges). These concrete bridges seemed to fail in the concrete. Note the crush forces and cracking of the I-880 pilings allowing some spans to fall with some other spans that had actual concrete fractures with rebar intact.

The question is… Are concrete bridges more likely to fail in an earthquake than a steel bridge? In other words does steel handle the rocking, thrusting, sliping, etc of an earthquake better?

I was peripherally involved with a program to reinforce concrete bridge columns by wrapping them with carbon fiber pre-preg. Old style columns (i.e. built before the lessons learned from the 1971 Sylmar earthquake), had most of the rebar oriented vertically. During an earthquake, the columns would flex causing the concrete to crumble, and with nothing holding the concrete, the column would fail by buckling. Post Sylmar columns have at least as much rebar wound helically as vertically, which keeps the concrete under compression and thus allows the column to keep supporting the weight (though repairs/replacement might be necessary).

Steel, unlike concrete, has good tensile st

[blue_streak_1]

Post Sylmar columns have at least as much rebar wound helically as vertically, which keeps the concrete under compression and thus allows the column to keep supporting the weight (though repairs/replacement might be necessary).

Steel, unlike concrete, has good tensile strength as well as compressive strength (often stronger in tension than compression). Steel is also ductile, in that it can bend without losing strength.

(Bear in mind that I’m an EE not a CE…)

• Erik

Ok that makes some sense. The next question is many concrete bridges are using prestressed concrete be3ams or panel beams. Will they be able to hold up as well as columns? Do new pre-stressed concrete beams now have this same helically wound re-bar?

[erikem]

I would think that prestressed beams without some sort of containment (e.g. a lot of rebar holding the concrete together in directions perpendicular to the prestressing tenons) would be susceptible to the same sort of failure as the old style bridge columns…

• Erik

[narig01]

My 2 cents worth.1. The recent earthquake in Japan is one of the strongest in recorded history. 2.During the 1989 Loma Prieta earthquake PG&E cut a lot of power mains because they had reason to believe that they had multiple gas main leaks. This was a precaution to reduce the risk of fire. 3. In the 1994 northridge earthquake.The freeway bridge that collapsed was directly over the newly opened Metrolink line to Lancaster. That disaster was the precussor to much of Metrolinks success Rgds IGN

[Paul_D_North_Jr]

Paul: I’m not sure how to interpret your question about “major” railroad bridges, but an earthquake centered NE of Montreal in Quebec something like 15 years ago reportedly caused a bridge to shift. I heard no details apart from news media reports, but my impression was of a standard girder span that dislodged from its bearings. Assuming that was the case, repair should have been fairly straightforward and fast. It might have been rather more serious if a train had come along (whether trying to outrun a tsunami or just in ignorance).

John

By the term “major” railroad bridge in this context, I meant one of - say, 3 or more spans, 300 ft. or longer, 50 ft. or higher, over a river or gorge, etc. The key criteria is a bridge that is not easily replaced or repaired quickly - a standard girder span would not be one of those, though, as the restoration could have involved just picking it up with a crane and re-setting it back in place, etc.

From Section 3.3.5 – General Design Criteria, 3^rd paragraph, middle of pg. 83, of the recent book, Design of Modern Steel Railway Bridges, John F. Unsworth, P.E., CRC Press/ Taylor and Francis Group, LLC, 2010, ISBN 978-1-4200-8217-3 (hardback) (emphasis added – PDN):

[Dragoman]

I’m no engineer, but something I have observed in Northern California since the 1989 Loma Prieta earthquake – concrete pylons holding up highway & railway (BART) viaducts have been encased with what appear to be steel plates – not apparently strong enough to hold up the structure, but strong enough to hold the concrete in place, even if it crumbles.

[blue_streak_1]

I’m no engineer, but something I have observed in Northern California since the 1989 Loma Prieta earthquake – concrete pylons holding up highway & railway (BART) viaducts have been encased with what appear to be steel plates – not apparently strong enough to hold up the structure, but strong enough to hold the concrete in place, even if it crumbles.

Can you tell if these steel plates are contoured to the shape of pylons??

[Dragoman]

Absolutely – some are actually oval (in the horizontal plane) to tightly hug the pylon

There might be more info at Caltrans’ or BART’s websites (under “seismic retrofit” perhaps, since that 's how the signs around the ongoing projects describe them), but I haven’t had a chance to check today (for some unfathomable reason, both the people around me and my tax clients think that I should actually spend time some getting work done!).

[Paul_D_North_Jr]

Excellent suggestion. For an good intro to this - which saves me from a lot of thought and typing - see this Wikipedia article: http://en.wikipedia.org/wiki/Seismic_retrofit

For info on the concrete columns, see this portion, and esp. the photo of the “jacketed and grouted column” at the right: http://en.wikipedia.org/wiki/Seismic_retrofit#Reinforced_concrete_column_burst

For bridge and viaduct retrofits, see also: http://en.wikipedia.org/wiki/Seismic_retrofit#Viaducts

For some of the principles as applied - and problems as found in - residential situations, see: http://www.bayarearetrofit.com/Avoid/avoid.html

There’s nothing wrong with concrete per se in earthquakes - but because of its weight and brittleness, a lot of attention has to be paid to the details. For instance, at a ‘joint’ between a horizontal and vertical member, it’s usually best to either keep it as a simple gravity connection so that the horizontal member can flex a little bit with regard to the vertical - or else reinforce the heck out of the connection and make it a massive casting so that it won’t fracture under the impact loading of the earthquake. Actually, a better solution is to go with the flexible connection, but provide vertical “shear walls” elsewhere in the structure to resist the sudden and large horizontal loads, which are the primary challenge with earthquakes.

For example, notable architect Frank Lloyd Wright designed a mostly concrete hotel for Tokyo in the 1916 - 1923 time frame. It survived the 1923 Great Kanto earthquake - magnitude 8.3 - when almost nothing else around it did, although with some mino