4 ms·

Confronting New Madrid, Part 2

23 points by ca98am79 11y ago

mcguire 11y ago
The article includes the following statement, which got my curiosity up: "It would shut down the Federal Express hub at Memphis and flatten large numbers of the rickety tilt-up warehouses, packed with inventory, that have grown up like mushrooms around the Memphis facility as companies try to out-compete each other in shipping speed." The word "tilt-up" was a link to Wikipedia[1], which article includes the section Risks: "In the wake of the 2011 Joplin tornado in which seven people were killed in a Home Depot when the 100,000 pounds (45 t) panel walls collapsed after the store was hit by an EF5 tornado, engineers in an article published in The Kansas City Star criticized the practice. They said that once one wall falls, it creates a domino effect. [...] Shortly after publication of the Kansas City Star article, the technical committee of the Tilt-Up Concrete Association (TCA) formed a task force to investigate the claims presented in the article. With the cooperation of Home Depot, the task group performed detailed engineering calculations, research and investigation of the claims posed in the article. [...] The final report was published on January 12, 2012.[22]" Unfortunately, the Wikipedia article doesn't say what the final report says. Fortunately, reference 22 has a link[2], and the report's executive summary has: "The Task Force’s findings to date include: "1. The failure started in the structural steel, steel joist and wide-rib deck roof system. This roof system is one of the most commonly used systems in commercial buildings, including those with masonry walls, precast concrete walls, and almost all forms of wall construction "2. The Tilt-Up concrete panels performed very well and survived the extreme loads of the EF-5 event only to collapse after the roof failed due to lack of bracing mechanism. "3. Tilt-Up construction methods played no role in the failure. "4. The perception that the nearby Wal-Mart store performed better because it was concrete masonry is false. The Wal-Mart took a glancing blow from the storm and the Home Depot took a direct hit." In other words, the roof of a large, open building is the weak link. On the other hand, as a professional programmer, I wonder how well typical commercial software development effort would perform under similar scrutiny. I know I haven't ever seen such an effort. [1] https://en.wikipedia.org/wiki/Tilt_up https://en.wikipedia.org/wiki/Tilt_up [2] http://tilt-up.org/tilt-uptoday/2011/12/19/task-force-completes-investigation/ http://tilt-up.org/tilt-uptoday/2011/12/19/task-force-comple...
- idlewords 11y ago
  You're comparing wind loads to seismic shaking, which stresses the structure differently. Phyllis explained to me that the problem with tilt-ups is lack of bracing strength - they're just not designed to be shaken back and forth for a long time. Compounding the problem is that the joints are very hard to inspect after the structure is built.
  - pronoiac 11y ago
    Hey, the author's here! Were you in the area when Iben Browning predicted a big earthquake from New Madrid, back in 1990? Did that pique your curiosity about it?
    - idlewords 11y ago
      I don't remember it firsthand, though it must have been in the news. I was 15 at the time and living near Chicago. Everyone on the trip was cursing the guy's name. I first got curious about New Madrid sometime in the 2000's after seeing a Wikipedia link, but don't remember the circumstances.
  - mcguire 11y ago
    I'm not actually comparing anything to anything; that was the only information on "rickety tilt-up warehouses" that I could find.
- jessaustin 11y ago
  There are forms of construction that don't require an intact roof to keep the walls upright. E.g., normal steel post-and-girder with regular wind frames/braces. Tilt-up just isn't such a form of construction. I'm surprised they can get away with this sort of building in Tornado Alley.
  - hga 11y ago
    Technically, we're several tens of miles east of Tornado Alley, but as I noted in my first comment, the accepted standard for residential and commercial building out here is 90 mph, which will handle more than 50% of all tornadoes (https://en.wikipedia.org/wiki/Enhanced_Fujita_scale https://en.wikipedia.org/wiki/Enhanced_Fujita_scale). And the odds of any particular location getting hit with any sort of tornado are small; in fact, I gather that derechos (https://en.wikipedia.org/wiki/Derecho https://en.wikipedia.org/wiki/Derecho) do more, but much less intense, damage since they effect vastly larger areas of land. It's just not worth the resources to build everything to handle > EF0 tornadoes, better to have alerting systems for people to listen to them and get to safety in time, although of course safety in an EF4/5 tornado is not trivial. We had 20 minutes of warning before it touched down, with 2 warnings sounded, and it took 5 minutes to make its way to my location. A large fraction of the fatalities were from people who simply didn't heed the warnings.

23 points by ca98am79 11y ago

The article includes the following statement, which got my curiosity up: "It would shut down the Federal Express hub at Memphis and flatten large numbers of the rickety tilt-up warehouses, packed with inventory, that have grown up like mushrooms around the Memphis facility as companies try to out-compete each other in shipping speed." The word "tilt-up" was a link to Wikipedia[1], which article includes the section Risks: "In the wake of the 2011 Joplin tornado in which seven people were killed in a Home Depot when the 100,000 pounds (45 t) panel walls collapsed after the store was hit by an EF5 tornado, engineers in an article published in The Kansas City Star criticized the practice. They said that once one wall falls, it creates a domino effect. [...] Shortly after publication of the Kansas City Star article, the technical committee of the Tilt-Up Concrete Association (TCA) formed a task force to investigate the claims presented in the article. With the cooperation of Home Depot, the task group performed detailed engineering calculations, research and investigation of the claims posed in the article. [...] The final report was published on January 12, 2012.[22]" Unfortunately, the Wikipedia article doesn't say what the final report says. Fortunately, reference 22 has a link[2], and the report's executive summary has: "The Task Force’s findings to date include: "1. The failure started in the structural steel, steel joist and wide-rib deck roof system. This roof system is one of the most commonly used systems in commercial buildings, including those with masonry walls, precast concrete walls, and almost all forms of wall construction "2. The Tilt-Up concrete panels performed very well and survived the extreme loads of the EF-5 event only to collapse after the roof failed due to lack of bracing mechanism. "3. Tilt-Up construction methods played no role in the failure. "4. The perception that the nearby Wal-Mart store performed better because it was concrete masonry is false. The Wal-Mart took a glancing blow from the storm and the Home Depot took a direct hit." In other words, the roof of a large, open building is the weak link. On the other hand, as a professional programmer, I wonder how well typical commercial software development effort would perform under similar scrutiny. I know I haven't ever seen such an effort. [1] https://en.wikipedia.org/wiki/Tilt_up https://en.wikipedia.org/wiki/Tilt_up [2] http://tilt-up.org/tilt-uptoday/2011/12/19/task-force-completes-investigation/ http://tilt-up.org/tilt-uptoday/2011/12/19/task-force-comple...

idlewords 11y ago
You're comparing wind loads to seismic shaking, which stresses the structure differently. Phyllis explained to me that the problem with tilt-ups is lack of bracing strength - they're just not designed to be shaken back and forth for a long time. Compounding the problem is that the joints are very hard to inspect after the structure is built.
- pronoiac 11y ago
  Hey, the author's here! Were you in the area when Iben Browning predicted a big earthquake from New Madrid, back in 1990? Did that pique your curiosity about it?
  - idlewords 11y ago
    I don't remember it firsthand, though it must have been in the news. I was 15 at the time and living near Chicago. Everyone on the trip was cursing the guy's name. I first got curious about New Madrid sometime in the 2000's after seeing a Wikipedia link, but don't remember the circumstances.
- mcguire 11y ago
  I'm not actually comparing anything to anything; that was the only information on "rickety tilt-up warehouses" that I could find.
jessaustin 11y ago
There are forms of construction that don't require an intact roof to keep the walls upright. E.g., normal steel post-and-girder with regular wind frames/braces. Tilt-up just isn't such a form of construction. I'm surprised they can get away with this sort of building in Tornado Alley.
- hga 11y ago
  Technically, we're several tens of miles east of Tornado Alley, but as I noted in my first comment, the accepted standard for residential and commercial building out here is 90 mph, which will handle more than 50% of all tornadoes (https://en.wikipedia.org/wiki/Enhanced_Fujita_scale https://en.wikipedia.org/wiki/Enhanced_Fujita_scale). And the odds of any particular location getting hit with any sort of tornado are small; in fact, I gather that derechos (https://en.wikipedia.org/wiki/Derecho https://en.wikipedia.org/wiki/Derecho) do more, but much less intense, damage since they effect vastly larger areas of land. It's just not worth the resources to build everything to handle > EF0 tornadoes, better to have alerting systems for people to listen to them and get to safety in time, although of course safety in an EF4/5 tornado is not trivial. We had 20 minutes of warning before it touched down, with 2 warnings sounded, and it took 5 minutes to make its way to my location. A large fraction of the fatalities were from people who simply didn't heed the warnings.