Any paleontologist will tell you that dinosaurs are extinct, killed off in the aftermath of what they believe was a massive meteor strike 66 million years ago.

The folks at the Insurance Institute for Highway Safety (IIHS), however, would beg to differ, and their proof is the collection of smashed cars displayed at their headquarters. Names like Chevrolet Venture and Mazda CX-9 may not evoke the kind of fear you’d feel coming face-to-face with a Tyrannosaurus rex, but once you’ve seen what these popular vehicles look like after the IIHS’s team of crash test engineers got their hands on them, you might think twice about going for a ride in one.

A non-profit funded by the American auto insurance industry, the IIHS dedicates much of its time and budget to evaluating the crashworthiness of the cars and trucks that millions of North Americans buckle themselves into each day. If you’ve ever seen the term “Top Safety Pick” referenced in a new car advertisement or heard it mentioned by a salesperson, that’s the label the IIHS bestows upon vehicles that best survive its harsh regime of crash tests.

Thanks to Subaru (a company eager to promote its status as the only automaker whose full lineup is recognized as Top Safety Picks), I was among the first group of Canadian writers to tour the IIHS’s vehicle research centre and witness firsthand the crashes, booms and bangs that happen at one of the most interesting workplaces in the auto industry.

Founded in 1959, the IIHS’s original purpose was to support other organizations’ vehicle safety research projects. About a decade later, it changed its focus from crash prevention to research on vehicle crashworthiness and how roadway designs contribute to driving-related deaths and injuries.

In 1992, the IIHS opened its vehicle research centre in rural Ruckersville, VA, and we’re not kidding about the rural part: the last 15 minutes of the bus ride from our hotel in nearby Charlottesville saw our driver dodge wayward cows before delivering us to the centre’s front door. Outside, it’s a peaceful spot, surrounded by swaying trees filled with singing birds. Inside, it’s a much different story.

In 2009, the IIHS celebrated its 50th birthday in a manner perfectly suited to a bunch of people who destroy stuff in the name of science: to demonstrate how far vehicle crash safety had come in the previous half-century, they lined up a pair of Chevrolet family sedans—a 2009 Malibu and a 1959 Impala – and crashed them into each other, head-on. The resulting carnage is the first thing you see when you walk in the front door, and if you think the video footage is dramatic, it pales compared to an up-close look at the cracked-up cars themselves.

We were treated to a comprehensive tour of the centre, but the highlight was watching a 2016 model (which we can’t name until the IIHS analyzes test data and compiles its results) put through the most recent addition to the crash-test roster, the small-overlap frontal test.

As one of two frontal crash evaluations the organization performs on every vehicle it rates, the small-overlap test sees the car hooked up to an under-floor propulsion system much like a slow-motion version of the catapults used to launch planes from an aircraft carrier’s deck. That system accelerates the car to 40 mph (about 60 km/h) and smashes the vehicle’s front corner, to a quarter of its width, into a fixed concrete barrier.

I was among a group of about 40 people standing on the viewing gallery above the crash hall as we watched the car approach down a long hallway, pass under us and hit the barrier, creating a shockwave we felt as much as heard. In moments it was all over: once the car stopped moving, the impact having spun it 90 degrees from its direction of travel, a crew of clean-up staff and engineers (including some from the automaker whose car was just destroyed) swarmed around it to clean up the debris and get a first look at how the car protected the crash-test dummy belted into the driver’s seat.

The IIHS is not the only U.S.-based vehicle safety testing organization. The government-run, publicly-funded National Highway Traffic Safety Administration (NHTSA) does similar work, but the IIHS’s (presumably) deeper, privately-funded pockets have allowed them to create a more challenging set of tests that they believe paint a more realistic crash safety picture.

“The (side-impact) barrier we developed (in the early 2000s) is based off measurements taken from the 19 biggest-selling pickups and SUVs (of that time),” said Raul Arbelaez. Today, he’s the IIHS’s VP of operations, but when he joined the organization in 1999 was one of the research engineers who developed the organization’s side impact test. “We didn’t like the barrier the government had; it was based off a (1970s-era) U.S. sedan, which was very low to the ground, and flat, and that meant the vehicle being hit was being loaded low on the rocker panel, which is a strong portion of the car.”

He’s also critical of the dummies the NHTSA was using in its side crash tests at the time. “They were not very good,” he said bluntly. “They weren’t bio-fidelic, so when you hit ’em, they weren’t going to tell you what a human would in terms of head and neck and pelvic injuries. The dummies we put in do all of that.”

They may be dummies, but they’re not cheap: basic models start at $40,000, rising to a quarter of a million bucks for those with data-capturing electronics inside. A newly developed side crash dummy is worth $660,000; needless to say, the IIHS hasn’t ponied up for any of those yet.

Not surprisingly, the dummies are not single-use throwaways: Arbelaez explained that while they do sustain damage during tests, parts are replaceable to save costs. “Ribs” in the torso often break during the side crash test, but cost “only” $30,000 for a full set.

After the test, the dummy’s data and video footage are uploaded wirelessly to the organization’s computer servers. It used to take weeks to deliver results to a carmaker, but modern technology means the manufacturer’s engineers walk away with gigabytes of data and video on a USB stick. “They leave here knowing more or less everything we know about the crash,” says Arbelaez.

Most recently, the IIHS added a test that eliminates the need for dummies altogether. Launched in 2013, the front crash prevention rating evaluates the effectiveness of electronic collision avoidance systems. Often using the same sensors that allow for ‘adaptive’ cruise control, slowing your car automatically when the vehicle in front decelerates, they provide automatic braking response in the case of an inattentive driver who doesn’t notice traffic has slowed significantly or stopped altogether. Where cars that previously performed best in the crash tests earned a Top Safety Pick rating, vehicles whose crash prevention systems also work well are now labelled Top Safety Pick+.

Because these tests require space for the vehicle to be driven at speeds up to 40 km/h to test how well collision avoidance systems work, the IIHS is building a covered (but not enclosed) outdoor ‘arena’ to allow for year-round avoidance testing.

Back to those vehicular “dinosaurs” I mentioned up top. Not all of them are decades old, like the ’50s Impala in the foyer. Mounted in the display hall like hunter’s trophies are some of the vehicles that have performed best and worst in more recent IIHS testing.

In the side impact test, added to the IIHS program in 2003, a stationary car is struck by a truck-shaped barrier, which is mounted on a sled that’s accelerated to 30 mph, or about 50 km/h. What’s left of a 2005 Mitsubishi Lancer shows what a poor side impact test result looks like: the B-pillar (the post that separates the sedan’s front and rear door openings) was severed where it used to meet the door sill, and pushed the driver’s seat way out of position.

“(This is) the only B-pillar that has ever split,” said Arbelaez.

Not far away hangs a post-test Chevrolet Venture, a popular minivan produced between 1997 and 2004. Its front-end structure essentially collapsed in the moderate-overlap front crash test: the A-pillar buckled, and the dummy was sandwiched between the seat and dash.

When Chevrolet redesigned the Venture into the Uplander in the mid-2000s, its engineers made improvements that earned the newer van a “good” rating in the same test. It was a similar story for the 2008 Lancer, a design that replaced the version that fared so poorly in that 2005 side test.

“(The 2008 Lancer) was a great performer with great structure and side airbags for front occupants, which shows these manufacturers made changes quickly; in an affordable vehicle, you had top-level protection,” said Arbelaez. He added that after the 2008 Lancer was tested, Mitsubishi’s engineers asked when the IIHS planned to take the old car down.

“No, we’re not going to do that. It’s too good a reminder of where things have been, and how far manufacturers have come.”

While automakers do adapt their designs to meet standards set by the IIHS tests, the organization continually strives to find new ways to challenge manufacturers’ work, a philosophy that led to the development of the small-overlap frontal test added in 2012.

Both moderate- and small-overlap tests are conducted with the car moving at about 60 km/h; the difference is in how much of the car’s frontal structure is involved in the impact. In the moderate-overlap test that made such a mess of the Chevy Venture, the car hits the barrier with half its width, but the small-overlap takes that down to just 25 percent.

“The outer 25 percent tends to have very little structure,” explained Arbelaez. “(It’s mostly) the steering components and wheel, and when you have these small-overlap crashes, it pushes that hinge pillar and can collapse the occupant compartment.”

The Mazda CX-9 is a big, seven-seat crossover, a type of vehicle that’s become popular with families, and based on its bulk, you’d think it would fare well in any crash. Not here: it earned a ‘poor’ rating in the small-overlap test. The obvious question is: what’s the difference between it and the Chevrolet Equinox displayed in the photo next to it, which earned a ‘good’ rating in the same test?

“For this Chev, one of the strategies they put in place was not only strengthening the occupant compartment, but at the end of frame rail (at the front corner, behind the bumper cover), they added some structure to deflect the crash forces and push some of that energy into the car’s longitudinal structure,” Arbelaez explained.

Arbelaez acknowledges the IIHS’s work isn’t perfect, agreeing with the suggestion that some crashes happen at speeds much higher than those replicated in testing.

“(Those speeds) don’t seem like much, but if you look at your typical T-bone crash, (50 km/h) can do a lot of damage, because you don’t have a lot of space between that intruding door and where you’re sitting.”

Ideally, Arbelaez says, the IIHS would crash every vehicle starting from low speeds, and ramp up to speeds that would lead to more severe crashes, but that would drive up costs: the IIHS purchases every vehicle it tests from dealerships to ensure carmakers don’t supply them with modified “ringers.” As it stands now, a car subjected to two frontal crashes, a side impact, and roof-strength testing requires four separate examples of any given vehicle.

“We do have higher speed limits (than we used to), but even when crashes happen at (100 or 130 km/h), often there’s speed scrubbed off in braking or running off road,” said Arbelaez. “Absolutely, some crashes happen at higher speeds, but when you look at the distribution of crashes, the majority are actually happening at (30 km/h) or less. If we had a test program that said every vehicle is poor at (160 km/h), it doesn’t really tell the consumer all that much when they go out and shop for a vehicle.”

There’s also the matter of collisions involving cars of different sizes. When gas prices rise, so do sales of smaller cars, and while the IIHS has no official test to gauge how small vehicles fare in collisions with larger cars, they did conduct unofficial simulations that pitted a manufacturer’s smallest car against popular mid-size models.

“The heavier object is always going to win,” said Arbelaez of the results. “The Smart Fortwo did well on its own, but against a Mercedes-Benz C-Class, its structure collapsed in a very different way.”

As well, that side impact test barrier is now based on truck and SUV designs from 15 years ago, while those vehicles have continued to grow larger to meet customer demand for aggressive styling.

“That is something we’ve talked about re-evaluating, but right now it seems to be doing a good job of driving improvements to structure and airbags,” said Arbelaez. “We are considering some other changes to the side impact program, but that’s more of a long-term thing for us.”

More Related Articles

Despite his reservations about the NHTSA’s test program, Arbelaez stressed that the key to finding a safe car is to choose one that does well in both organizations’ tests. IIHS research shows that an improvement from ‘poor’ to ‘good’ in IIHS side-impact testing correlates to a nearly 50 percent reduction in fatality rates, and that a car with a strong roof structure also significantly reduces rollover deaths.

“What we’ve done is look at real-world data, and what the feds were doing (to evaluate crash safety), and where there were deficiencies, we wanted to see what we could do,” said Arbelaez.

Change may not happen all at once like it did following that meteor strike millions of years ago, but the IIHS’s work continues to have a positive impact on the number of modern-day dinosaurs found on our roads today.

Connect with Autos.ca