Five people were killed and at least 45 concert-goers injured when a scaffolding blew over in 70 mph winds at the Indiana State Fair this weekend.
Audience video captures the horror:
Already, engineers are buzzing over how this tragedy could have been prevented. Such scaffoldings are commonplace and designed, apparently, to withstand gusts up to 90 mph. There are always tradeoffs between weight, costs, safety and usability.
And as investigators sift through the wreckage, clues could emerge suggesting defective joints, missing bolts, fatigued metal or other possible causes.
If you were tasked with developing the next-generation scaffolding that needed to stand up in hurricane-force winds, how would you do it in a practical, cost-effective way?
Dashpots, springs, and weights can be modeled electrically. Can a truss be modeled electrically?
To what extent could a structure like this be scaled down and subjected to wind tunnel tests? Also, with the continued miniaturization of sensors, it seems with some structures you might be able to create a network of stress sensors and start taking data on the forces they are subjected to.
Given the exposed nature of the stage at the fair it would be hard to protect it and fair goers from injury/damages. I wonder what type of connection with the ground was made? What if the structure was reinforced with strong guy wires that both held it down in place and prevented side to side swaying? I was wondering what (if any) covering was on the top structure of the staging; if it looked like an airfoil to the winds then there could be a unintended set of stresses on the structure. Given the temporary nature and the possible onsite setup/assembly it is possible that the best engineered structure would not survive high winds (if for example the right number and type of connectors were not used). Just wondering, my heart goes out to those who lost loved ones and to those recovering.
It was reported that the NWS issued a high wind warning only 10 minutes before the disaster. In my part of the country these warnings are typically not issued until the weather event is directly overhead. Maybe a little less reliance on computer models and more human judgement at the NWS offices could get these warnings out sooner.
I think that is an unreasonable requirement or expectation for temporary structures. Mitigation through evacuation. We'll have to see what the investigation reveals but one change that might have prevented the collapse would be to ensure the covering would "sacrifice" itself and fly off to reduce the wind loads on the rest of the structure. Unfortunately, the engineer loses control of the structure once it leaves the factory and there are any number of (unapproved or unintended) things that the end users will do with these. Given enough time or money you can do almost anything. Question is, will someone want to pay for it?
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.