Accident Re-Creation
It would be wonderful if we can simply rewind life's tape a few minutes and review how exactly an accident happened. Actually, the chances of having a record of accidents are increasing because more and more CCTV cameras are being deployed. Also, with the ever-increasing versatility and power of hand-phones, some might record the accident 'accidentally' while they were shooting something else, or purposely because they just happened to be there with a camera on stand-by mode.
In the West, there is money to be made from such records, by selling them to the media, or to one side or the other (or both!), in the subsequent investigation - East may not be slow to catch up!
Barring such direct records, the next best thing is a re-creation (surely not 'recreation'!) of the accident - also called 're-enactment' or 'reconstruction'. This can get pretty complicated.
For one thing, things are never the same at a microscopic or instantaneous level. Most materials may be reproducible, but the exact sequence and intensity of actions may not be. For another, some acts are not feasible to reproduce, such as a live person falling off from height. A cardinal rule in re-creation is that no human, and nowadays, no animals either, may be harmed, and no irreplaceable or expensive damage occur in the process of re-creation. Still, a lot of information can be gleaned from an intelligent re-enactment of the accident.
Here, scale becomes a problem. For instance, cell-phone and auto makers, even aircraft manufacturers, will be happy to sacrifice one or more of their products to undergo various destructive tests. But a civil engineer, involved with huge projects, generally does not have the luxury. He can only put a small segment of the structure through the accident scenario.
In citing examples of this, I would like to start with a non-civil engineering example of how sometimes the simplest and the cheapest of tests can prove a major truth.
NASA Challenger Disaster
During the heyday of American space supremacy, sending crews into space orbit for various lengths of time to carry out experiments and bringing them back to earth became a routine chore like sending kids to school and receiving them back in the evening. The Space Shuttle program became a bus ride.
It was during the tenth mission on 28 January 1986, launched in very cold weather against the warnings of some of the engineers involved, at about 73 seconds after launch, the shuttle blew up, and the pieces fell into the ocean, killing all the seven crew members, (Figure 16.)
The culprit was a gasket called the 'O-ring', which lost its elasticity in the extreme cold and failed to seal the hot combustion gases from cabin crew and instrumentation. The web is replete with details of the disaster, starting with a brief account from Wikipedia.
The anecdote I wish to relate here is the way the gasket's dangerous behaviour in cold temperature was demonstrated by Nobel Prize winner Richard Feynman, while giving expert evidence on NASA's Challenger failure (Fig. 16, Right).
To prove how easy it was to understand this, Feynman asked for a glass of ice-water while waiting for his testimony for the Presidential Commission, and soaked a sample of the material of the 'O-ring' compressed in a 'C-clamp' which he had picked up on the way from a hardware store. When his turn came to speak, he simply pulled the sample out of the water, and showed that when he removed the clamp, the specimen did not spring back to its original shape, having lost its elasticity temporarily, destroying the integrity of the 'O-ring' seal!
Fig. 16. The Challenger Disaster. Left - Launch, crew of 7, and blow-up. Top - Richard Feynman shows cold effect on gasket
If this sounds like something out of the American show CSI or Perry Mason stories by Erle Stanley Gardner, I can confirm that such is possible in the USA. Even in other countries, as long as you understand and abide by the local culture, you can present your re-enactment results with telling effect.
Field Test for Support Rotational Capacity
It may be sacrilege to speak about my Mickey Mouse experiment in the same breath as Feynman’s demo above, but it will illustrate the point that sometimes a simple field test may prove an important valid point.
In the bar-chair system collapse case described earlier, one of my contentions to explain the collapse was that the vertical supports had very little lateral resistance. The defence had conducted lab tests to document its tension, compression, and shear tests, but had overlooked the rotational capacity. I did not have the time (or the means!) to conduct a similar lab test for rotation, so I chose to set up my own ‘quick-and-dirty’ field test.
I had my lawyers arrange with a local contractor to set up a single bar of the same size as the leg of the collapsed bar-chair with same end conditions. I got a worker pull gradually at top with a rope, with the pull force being measured by a simple spring balance, (Figure 17.)
Fig. 17. Author’s field test for rotational capacity
When I presented this in court, the defence lawyers argued at my conducting such an unprofessional test which would prove nothing of value in the case. Briefed by their expert, they asked if I had calibrated the spring balance, which was a common 25kg capacity household item bought from a local store. I explained that I had not wanted to conduct a high-tech lab test to fractions of a kg force, and was willing to accept the usual precision of say plus or minus 1kg for the household spring, which would be sufficient for my purpose.
After all, the bar fell at about 15kg pull, which was a small fraction of what would have been necessary to prevent collapse of rebar cage under lateral force. What difference would it have made if it had been 16kg, or even 18kg? It confirmed my argument that rotational mode of failure was much easier than any other mode, and that was the way the cage had failed.
The defence had also conducted a re-creation of the accident in a university lab taking a sizable segment of the bar-chair and top rebar system, at considerably greater time and expense. On that, I presented my own arguments as to how a too-well organised and planned lab test might not explain the collapse due to site deficiencies - but that is another story.
None of the arguments from both sides was really conclusive, because there were various other factors at work contributing to the accident. But I had my two-cent's worth at much less expense than the other side!
Re-creation of Ladder Accident
When someone gets hurt while using a commercial product, quite often (particularly in the West), product liability claims can run into the millions, and then, experts would have an awesome job to prove 'beyond reasonable doubt', the culpability of the manufacturer or facilitator of the product.
A case which illustrates such a re-creation is the forensic analysis of a fall from a ladder in UK in the 1990s. The 73kg owner of a house cleaning the second floor window from his recently bought extensible ladder fell while climbing. He approached the Legal Aid Board for an opinion on whether he can sue the ladder company for a faulty product.
One of the methods the Board’s expert used was to re-create the fall using a weight suspended from different rungs of the ladder corresponding to different positions of the user, and also for different angles of the ladder with the horizontal, (Figure 18.)
It was established from the owner's description of how he had set up the ladder that the owner had started with an angle of ladder about 56° with the horizontal as against the approximately 75°
Fig. 18. Re-creation of ladder fall (Far ladder is for control.)
Tests with the dummy weight demonstrated that when the weight was hung near the top, as the owner was positioned when he fell, at the 56° angle of the ladder, it became unstable and the foot slipped away from the wall, the top sliding down the wall. In the right portion of Fig. 18, the ladder is static only because the top has been stopped by a ledge on the wall.
As the owner had not followed the manufacturer’s instructions for slope, he would not be able to sue them, according to the Legal Aid Board. Had it been otherwise, in such product liability cases, the payout tends to be huge (often in the millions!) in Western countries.
Rest of the Forensic Analysis:
A side-note to this example is worth adding. The Board investigation conducted so many other analyses and tests as follows, with reference to Fig. 19:
- Ladder draw-down marks on the wall;
- Impact damage to the sill above patio door;
- (a),(b)Ladder tip fracture surfaces, (c)Failed ladder tip, abrasion from brick;
- Infrared spectrum of ladder tip material;
- Uncontrolled side slip marks on wall (on white board tacked for that purpose);
- Melting curve of Differential Scanning Calorimetry (DSC) for tip material;
- Coefficient of Friction vs. Load, for ladder tip material;
- Foot-prints of ladder feet at 75° and 60°; and,
- (a),(b)Various statics analyses of ladder friction on surfaces of different slopes.
This is an unusually comprehensive list of forensic investigations for a ladder accident - but it shows how much a serious investigation would involve!
Fig. 19. Various tests and analyses run in ladder investigation