Volvo's crash test mummies

By DAVID LINKLATER

Volvo has long been acknowledged as a carmaker that looks after family safety. But now the Swedish firm is taking its research one step further with the creation of the world's first official computer model of a pregnant crash test dummy.

The mother-and-baby model is significant because despite the billions of dollars spent each year by the automotive industry on crash testing, little is known about how accidents affect unborn children.

"Now we've covered the whole lifecycle," says Camilla Palmertz, a biomechanical engineer at the Volvo Cars Safety Centre. "One big advantage is that both mother and baby can be scaled up or down to the size we want to study."

This virtual crash dummy is a woman at a late stage of her pregnancy, since that is when the unborn baby is at greatest risk in an accident. The basic geometry had been completed by January. Since then, the centre has been refining the model.

"Now it's finished and we've started running simulated front-end impact tests on it."

The purpose of the tests is to study how the virtual mother-to-be and her unborn baby are affected by the seat belt and airbag in simulated accidents. The computer model makes it possible to study - in great detail - how the belt moves, the influence of belt and airbag on the uterus, placenta and foetus, and how the foetus moves in relation to the mother's body. The model can also be used to test new designs for seat belts and other safety systems.

"I'm certain that there's room for further development of the three-point belt, to make it more comfortable and to provide even better protection," says Palmertz.

Many pregnant women wonder whether seat belts could harm their unborn babies if they are involved in an accident.

A recent study in the field of biomechanics at the University of Halmstad, Sweden, found that 46 per cent of pregnant women complained of belt chafing against the abdomen, 10 per cent said that the belt was too tight against the breasts, and 7 per cent found that the belt was not long enough.

Seven per cent also admitted to stopping seatbelt use because of comfort problems. All researchers agree that belts should be worn - but in the right way, explains Palmertz.

"It should be between the breasts and as low as possible over the hips. The lap section of the belt mustn't be allowed to ride up in front of the woman's tummy."

In an accident, the pregnant woman's thorax and pelvis are both restrained by the belt, but her abdomen is free to move in the direction determined by the particular forces arising from the impact. Because the foetus is floating free inside her, injuries tend to fall into two main types. The more common of the two is that the placenta becomes either partially or completely detached, which means that the baby can't get enough oxygen. The rarer scenario is for the head of the baby to be injured if it hits one of the bones of the mother's pelvis.

"We think that the placenta may become detached. While the uterus is relatively elastic and can therefore change shape, the placenta is not equally resilient under acceleration," says Palmertz.

"We think so, but we don't know yet. That's why this model is so important."

Most of the crash tests that take place at Volvo's Safety Centre near Gothenburg, Sweden, involve cars and people that don't exist. The centre's NEC SX-4 "supercomputer" stands in a climate-controlled room and can run the same accident scenario several times at different speeds, with different safety systems and with different occupants. Six simulated full-car crashes can be carried out over 24 hours.

While virtual testing yields crucial data in the early stages of product development, only a physical crash test can account for all variables in an accident.

The centre also has a moveable track suspended on air cushions that can be turned by up to 90 degrees. This allows car-to-car impacts from full-frontal to right-angle side impacts.