torsion

March 11, 2016 § Leave a comment

A great few days last wee reviewing projects at Cal Poly and talking to this year’s docent class at the Chicago Architecture Foundation.  Anyone who knows me knows that those are two almost sacred places–both of them devoted to bringing art and science together in how people design and understand architecture.  The weather was a bit better in California (as was the wine tasting), but the welcome was much appreciated all around.

Figure 09

Explaining the complexity of the Monadnock to docents has been one of the happy goals of my participation in their education program.  It’s a good chance to talk about the stories that we tell about these buildings and the almost impossible richness that lies under the skin–literally.  So while it’s fine for the overview to talk about the building being “the last of the bearing wall skyscrapers,” I enjoy the chance to pull out this image and to talk about how it’s really a hybrid that incorporates brick piers, brick shear walls, steel columns, steel portal frames, structural masonry and cladding masonry, etc., etc.  And how John Root’s choices in detailing very cleverly make us read the building as bearing wall, but how in fact the bay windows (obviously, when you think about it) can’t possibly be bearing walls.

I’ve droned on about this aplenty before, but this time around one docent came up to me at the break and asked about those steel portal frames and why you’d need them.  Was it, he asked, so the building didn’t twist?  And the answer is–absolutely right.  In structures classes, we talk about the three ways buildings can fail.  They can 1) fall down (gravity loads), or 2) fall over (wind loads), or they can 3)…uh…rotationally pancake.  This latter failure is a failure in torsion, and it can happen if the structure has enough lateral resistance perpendicular to a wind (or seismic, or impact) load, but unequally distributed.

torsion

So, in the floor plate on the left, for instance, the shear walls are concentrated on one side of the structure, making that side much stronger than the other.  There might be enough shear resistance in those walls to keep the building from falling over, but the ‘weak arm’ of the building will deflect more, and as it spins around the center of resistance, the columns that hold it up will necessarily rack, while the shear walls will stay in roughly the same spot.  As the columns rack, the floor will have to drop at that end, and eventually the entire thing can twist and flatten.  The solution is to add shear resistance to make sure that the entire building has approximately the same resistance to wind, seismic, etc., across its floor plates, and the addition of a shear wall to the floor plate on the right does this.

plan

In the Monadnock, the north end of the floor plate (for whatever reason, this seems stuck at the bottom of the image–sorry on behalf of WordPress’ editing software), probably for reasons of efficiency, doesn’t have a masonry shear wall like the south end of the original.  Instead, Burnham and Root designed a portal frame–basically a deep truss girder that fixes the angle between column and girder at 90°.  For this end of the building to rack, the wind is going to have to actually deform the steel column and girder.

There are, of course, other examples of torsion as a design problem nearby.  While most of the buildings along Dearborn–the most famous street in lateral resistance history–are symmetrical, there’s one that’s decidedly not symmetrical.  Inland Steel gains much of its lateral resistance in the east/west direction from its separated core, which is nudged to the southern end of its floor plate.  That leaves the north end as a potential weak story (not to mention a completely illegal fire exiting situation).  One of Fazlur Khan’s first tasks as a young engineer at SOM was to figure out how to distribute enough lateral resistance throughout Inland’s frame to keep the north end from twisting around its core.  His solution was to distribute much of the necessary lateral resistance among all of the building’s girder/column joints, in a connection called a torque box.  By welding additional plates in the interface between the column and girder, Khan was able to guarantee the 90° relationship between elements throughout the frame, making the north end stiff enough to deflect more or less in tune with the core to the south.

 

inland plan

A million thanks, as always, to Jennifer Masengarb and Hallie Rosen at CAF for inviting me and for organizing a great morning.  Always honored to be able to contribute to such a great program!

torque box

Model by Doug Conroy and Asa Westphal

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

What’s this?

You are currently reading torsion at architecturefarm.

meta