This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Interview: The human crash test dummy
05 May 2007
From New Scientist Print Edition.
Justin Mullins

How did you become a human crash test dummy?

By accident. Sorry, you asked for that. I was a policeman in California teaching accident investigation. I needed a way to do some hands-on experiments. The only way I could do that was to get in a car and crash it.

Why are artificial crash test dummies of such limited use?

They are designed for high-speed crashes. To allow them to be used many times, they have to be built to last - for example, the neck is very stiff to prevent it breaking. That extra stiffness makes no difference to the way the dummy reacts in a high-speed crash, but at lower speeds it means the dummy responds differently from a real person.

How do you measure the effects of a car crash?

We use accelerometers on the car and its occupants to measure the change in velocity. Then we look at data from other studies that have analysed the injuries people have suffered in crashes. We can then make predictions about the injury potential of crashes.

How do you categorise crashes?

We describe the potential severity of a crash not in terms of the speed at which the impact occurs, which can be misleading, but of the change in speed. We call this the delta-v. A change in speed of less than 16 kilometres per hour is a low-speed crash. Between 16 and 40 kilometres per hour, you're in the moderate range, while 40 to 55 kilometres per hour is moderate to severe.

What is the highest delta-v you've personally have experienced?

43 kilometres per hour.

What kind of force do you face in test crashes?

Normally a little more than 10g on average - that's 10 times the force of gravity - and that lasts for just over a 10th of a second. So, say I weigh 200 pounds, for a brief moment in time it feels as if I weigh 2000 pounds. That's why you cannot really brace against a crash.

The force of the crash can put a lot of stress on my ribs and I do run the risk of breaking them. Beyond that, biomechanics experts talk about the "third collision". The first collision is a car hitting another car and the second is the passenger's body hitting the dashboard or steering wheel. The third is when internal organs such as the heart or the brain hit something else inside the body, such as ribs or the inside of the skull. I have never been injured, though, because I usually do lower-speed crashes.

How sore do you get?

One day last year in Las Vegas I did nine crashes between 8 in the morning and noon. Some of them involved quite substantial delta-vs. That evening, my ribs and shoulders were certainly sore and I was tired, but it's hard to tell how much of that was from pushing cars around and from lifting this and carrying that.

Are low-speed crashes the most common?

Yes they are, but a more interesting question is how much injury they really cause. The answer depends on what society you live in. In the US, we have a high level of reported injuries from low-speed crashes and we also have a lot of litigation. In countries where there isn't that level of litigation, you don't find so many reported injuries. It's a tough one to answer.

What is the safest kind of car to drive in?

Bigger is better. When two cars collide, the one with the greater mass will experience a smaller delta-v and so should be safer for its occupants. Of course, the safest cars are also the ones with the most up-to-date safety features. If everyone drove the same-sized car, it would be a lot easier to say which other features of cars make them safer.

Where do you stand in the debate on whether cars should carry black boxes to collect data, like aircraft do?

I am totally in favour of it. People worry that it would invade their privacy. They are envisaging something that tracks where they go and who they're with, but these systems don't do that. The data recorder only records information when there is a crash. That is not a privacy violation because the information is going to be discovered anyway through an accident investigation, by measuring the tyre marks on the road, for example. It records no personal information at all, so how can there be a privacy issue? I think this is a meaningful use of technology to save lives.

How does your work make the roads safer?

What we do is teach people how to investigate car wrecks. If there are 50 crashes at a particular junction and you don't know what causes them, what can you do to prevent the 51st? You first need to investigate those 50 wrecks and identify a common factor. Suppose, for example, we find that 45 per cent of them were caused by people driving too fast southbound, what can we do to stop that? Traffic safety is all about the three Es: engineering, enforcement and education. We can re-engineer the road so that drivers cannot go so fast, change the speed limit or put in a stop light. Or we can educate folks by putting up billboards telling them to reduce their speed. Or we could use enforcement: we could ask the police to give out tickets for speeding. But none of that is possible unless you've investigated the crashes.

From issue 2602 of New Scientist magazine, 05 May 2007, page 50-51

Profile Rusty Haight is director of the Collision Safety Institute, a crash research, training and consulting centre based in San Diego, California. He holds the Guinness World Record for the number of crash tests he's taken part in as a human dummy, currently 846. In 2004, Men's Journal magazine declared him the 24th toughest guy in the US.

For the latest from New Scientiist visit www.newscientist.com