Versaille Wedding Hall disaster

This is a public blog posting of a structural engineering discussion thread on a forum I post on. I figured that since I was spending the time to write up these pieces that I may as well post them outside that insular forum and make these educational posts public.

This post will be about the Versaille wedding hall disaster if you hadn't guessed already.
http://en.wikipedia.org/wiki/Versailles_wedding_hall_disaster

This structural collapse is also notable for being caught on camcorder:

The wiki explains it quite well, but this case is notable because it could have easily been avoided. One should note that structural failures often occur because many things fail, and it is often ignorance which leads to catastrophe. Although the building in this case was not correctly built/designed, it still would not have failed if there were not even further errors down the line. This is fairly typical of massive structural failures - it is often a combination of many things going wrong that cause a structure to fail, and even an incorrectly constructed building is far tougher than most will give it credit for (but not always).


So what went wrong?



First lets look at what happened during construction:

Here is the original design of the building, the floor which later collapsed is marked in red. Note that it is actually a roof in this drawing - the floor was originally intended to be the rooftop.

Late in construction, they decided to add in an extra floor. Hence the new roof structure built on top. This is not unusual in of itself, although the new floor must be designed to carry a heavier load. Wiki says that this re-design may not have happened. I should note here that standard floor loading where people gather (such as this wedding hall) is roughly double that of a traffic-able roof.

As it turns out, the new floor seemed to function ok. A few weeks before the disaster the owners decided to remove the partitions in the floor below. This now created a problem, as the overloaded floor above (red line) was actually being supported by those partitions. Depending on how a structure is designed its very common to have a floor slab not rely on internal partitions. The overloaded floor slab over was in fact bending too much and was actually resting on these partitions - which is why it appeared to be function ok. These partitions then spread the load onto the next floor. This leads to the next image.

After the partitions below had been removed the floor had started to sag. It was at this point that the owners should have called in an expert and disaster would have been averted. Instead they noticed that the floor was sagging, and they used screed to level off the floor - which increased the load even more.

This structural failure was caused by poor building management practices. This case is also notable in that an engineer who was not involved in the design or construction of this building went to jail for developing the method of construction used. The take-away from this is that if you notice a floor, or beam, or any other structural system showing signs of distress (excessive sagging, cracking, deformation) you should let someone know. Minor cracking in plaster or brickwork is normal though so don't panic about that.