• Apr 14th 2009 at 12:23AM
  • 109
IIHS crash tests – Click above for high-res image gallery

One of the biggest concerns people have with buying smal cars is crash safety. More specifically, there are concerns about what happens when a small car meets a larger one in an accident. Regardless of size, all cars are required to meet the same crash safety standards. The problem is that crash tests are done by slamming a car into a standardized barrier, while accidents in the real world don't generally involve impacts with standardized barriers.

When cars of different sizes collide there are a huge number of variables that impact the result. One of those is kinetic energy, which is a function of mass and velocity. A larger, heavier vehicle has more energy than a smaller, lighter one, and in combination with less crush space, the latter tends to come off worse in an accident. The Insurance institute for Highway Safety has run a series of crash tests between sub-compact and mid-sized cars and in all three cases the small cars did substantially worse than they did in the standard crash tests. All of the small cars earned good scores in regular testing but did poorly in the new tests. The IIHS press release explains the results after the jump.

  • Photo from the Institute's 40 mph frontal offset offset crash of the Yaris into a deformable barrier: The dummy's position in relation to the steering wheel and instrument panel after the crash test indicates that the driver's survival space was maintained well. The Yaris earned a GOOD rating in this test. (Tested vehicle: 2007 Toyota Yaris)
  • Photo from the Institute's 40 mph offset frontal crash of the Fit into a deformable barrier: The dummy's position in relation to the steering wheel and instrument panel after the crash test indicates that the driver's survival space was maintained well. The Fit earned a GOOD rating in this test.
  • Camry pre-crash photo
  • Yaris pre-crash photo
  • The Camry earned an ACCEPTABLE rating in this test; the Yaris earned a POOR rating.
  • Photo of the Yaris from the Institute's 40 mph frontal offset crash with the Camry: There was a lot of intrusion into the occupant compartment. The driver's seat tipped forward, and the steering wheel moved excessively. The head injury measure on the dummy rated poor, and there was extensive force on the neck and right leg plus a deep gash at the right knee.
  • Accord pre-crash photo
  • Fit pre-crash photo
  • The Accord earned a GOOD rating in this test; the Fit earned a POOR rating.
  • Photo of the Fit from the Institute's 40 mph frontal offset crash with the Accord: There was a lot of intrusion into the occupant compartment, which compromised the survival space around the driver dummy. Measures recorded on the dummy indicate that the risk of serious injury would be high in a similar real-world collision.
  • C class pre-crash photo
  • Smart Fortwo pre-crash photo
  • The C class earned a GOOD rating in this test; the Smart earned a POOR rating.
  • Photo of the Smart from the Institute's 40 mph frontal offset crash with the C class: The space around the driver collapsed during the crash. Multiple injuries, including to the head, would be likely for a real-world driver of a Smart in a similar collision.

[Source: Insurance Institute for Highway Safety]


New crash tests demonstrate the influence of vehicle size and weight on safety in crashes; results are relevant to fuel economy policies

ARLINGTON, VA - Three front-to-front crash tests, each involving a microcar or minicar into a midsize model from the same manufacturer, show how extra vehicle size and weight enhance occupant protection in collisions. These Insurance Institute for Highway Safety tests are about the physics of car crashes, which dictate that very small cars generally can't protect people in crashes as well as bigger, heavier models.

"There are good reasons people buy minicars," says Institute president Adrian Lund. "They're more affordable, and they use less gas. But the safety trade-offs are clear from our new tests. Equally clear are the implications when it comes to fuel economy. If automakers downsize cars so their fleets use less fuel, occupant safety will be compromised. However, there are ways to serve fuel economy and safety at the same time."

The Institute didn't choose SUVs or pickup trucks, or even large cars, to pair with the micro and minis in the new crash tests. The choice of midsize cars reveals how much influence some extra size and weight can have on crash outcomes. The Institute chose pairs of 2009 models from Daimler, Honda, and Toyota because these automakers have micro and mini models that earn good frontal crashworthiness ratings, based on the Institute's offset test into a deformable barrier. Researchers rated performance in the 40 mph car-to-car tests, like the front-into-barrier tests, based on measured intrusion into the occupant compartment, forces recorded on the driver dummy, and movement of the dummy during the impact.

Laws of physics prevail: The Honda Fit, Smart Fortwo, and Toyota Yaris are good performers in the Institute's frontal offset barrier test, but all three are poor performers in the frontal collisions with midsize cars. These results reflect the laws of the physical universe, specifically principles related to force and distance.

Although the physics of frontal car crashes usually are described in terms of what happens to the vehicles, injuries depend on the forces that act on the occupants, and these forces are affected by two key physical factors. One is the weight of a crashing vehicle, which determines how much its velocity will change during impact. The greater the change, the greater the forces on the people inside and the higher the injury risk. The second factor is vehicle size, specifically the distance from the front of a vehicle to its occupant compartment. The longer this is, the lower the forces on the occupants.

Size and weight affect injury likelihood in all kinds of crashes. In a collision involving two vehicles that differ in size and weight, the people in the smaller, lighter vehicle will be at a disadvantage. The bigger, heavier vehicle will push the smaller, lighter one backward during the impact. This means there will be less force on the occupants of the heavier vehicle and more on the people in the lighter vehicle. Greater force means greater risk, so the likelihood of injury goes up in the smaller, lighter vehicle.

Crash statistics confirm this. The death rate in 1-3-year-old minicars in multiple-vehicle crashes during 2007 was almost twice as high as the rate in very large cars.

"Though much safer than they were a few years ago, minicars as a group do a comparatively poor job of protecting people in crashes, simply because they're smaller and lighter," Lund says. "In collisions with bigger vehicles, the forces acting on the smaller ones are higher, and there's less distance from the front of a small car to the occupant compartment to 'ride down' the impact. These and other factors increase injury likelihood."

The death rate per million 1-3-year-old minis in single-vehicle crashes during 2007 was 35 compared with 11 per million for very large cars. Even in midsize cars, the death rate in single-vehicle crashes was 17 percent lower than in minicars. The lower death rate is because many objects that vehicles hit aren't solid, and vehicles that are big and heavy have a better chance of moving or deforming the objects they strike. This dissipates some of the energy of the impact.

Some proponents of mini and small cars claim they're as safe as bigger, heavier cars.

But the claims don't hold up. For example, there's a claim that the addition of safety features to the smallest cars in recent years reduces injury risk, and this is true as far as it goes. Airbags, advanced belts, electronic stability control, and other features are helping. They've been added to cars of all sizes, though, so the smallest cars still don't match the bigger cars in terms of occupant protection.

Would hazards be reduced if all passenger vehicles were as small as the smallest ones? This would help in vehicle-to-vehicle crashes, but occupants of smaller cars are at increased risk in all kinds of crashes, not just ones with heavier vehicles. Almost half of all crash deaths in minicars occur in single-vehicle crashes, and these deaths wouldn't be reduced if all cars became smaller and lighter. In fact, the result would be to afford less occupant protection fleetwide in single-vehicle crashes.

Yet another claim is that minicars are easier to maneuver, so their drivers can avoid crashes in the first place. Insurance claims experience says otherwise. The frequency of claims filed for crash damage is higher for mini 4-door cars than for midsize ones.

Here's how the pairs of cars fared in the Institute's new crash tests:

Honda Accord versus Fit: The structure of the Accord held up well in the crash test into the Fit, and all except one measure of injury likelihood recorded on the driver dummy's head, neck, chest, and both legs were good. In contrast, a number of injury measures on the dummy in the Fit were less than good. Forces on the left lower leg and right upper leg were in the marginal range, while the measure on the right tibia was poor. These indicate a high risk of leg injury in a real-world crash of similar severity. In addition, the dummy's head struck the steering wheel through the airbag. Intrusion into the Fit's occupant compartment was extensive. Overall, this minicar's rating is poor in the front-to-front crash, despite its good crashworthiness rating based on the Institute's frontal offset test into a deformable barrier. The Accord earns good ratings for performance in both tests.

Mercedes C class versus Smart Fortwo: After striking the front of the C class, the Smart went airborne and turned around 450 degrees. This contributed to excessive movement of the dummy during rebound - a dramatic indication of the Smart's poor performance but not the only one. There was extensive intrusion into the space around the dummy from head to feet. The instrument panel moved up and toward the dummy. The steering wheel was displaced upward. Multiple measures of injury likelihood, including those on the dummy's head, were poor, as were measures on both legs.

"The Smart is the smallest car we tested, so it's not surprising that its performance looked worse than the Fit's. Still both fall into the poor category, and it's hard to distinguish between poor and poorer," Lund says. "In both the Smart and Fit, occupants would be subject to high injury risk in crashes with heavier cars." In contrast, the C class held up well, with little to no intrusion into the occupant compartment. Nearly all measures of injury likelihood were in the good range.

Toyota Camry versus Yaris: There was far more intrusion into the occupant compartment of the Yaris than the Camry. The minicar's door was largely torn away. The driver seats in both cars tipped forward, but only in the Yaris did the steering wheel move excessively. Similar contrasts characterize the measures of injury likelihood recorded on the dummies. The heads of both struck the cars' steering wheels through the airbags, but only the head injury measure on the dummy in the Yaris rated poor. There was extensive force on the neck and right leg plus a deep gash at the right knee of the dummy in the minicar. Like the Smart and Fit, the Yaris earns an overall rating of poor in the car-to-car test. The Camry is acceptable.

Fuel economy implications: One reason people buy smaller cars is to conserve fuel. Gasoline prices skyrocketed last year, and there's no telling what the price at the pump might be next week. Meanwhile, the gears are turning to hike federal fuel economy requirements to address environmental concerns. The conflict is that smaller vehicles use less fuel but do a relatively poor job of protecting people in crashes, so fuel conservation policies have tended to conflict with motor vehicle safety policies.

A problem with the current structure of fuel economy standards for cars is that the target of 27.5 miles per gallon is applied to an automaker's whole fleet, no matter the mix of cars an individual automaker sells. This encourages manufacturers to sell more smaller, lighter cars to offset the fuel consumed by their bigger, heavier models. Sometimes automakers even sell the smaller - and less safe - cars at a loss to ensure compliance with fleetwide requirements.

In response, the Obama administration announced it is boosting the fuel economy standard for cars, beginning with 2011 models, and instituting a size-based system to set fuel economy targets like the one already in effect for SUVs, pickups, and vans. This system will mandate lower fuel consumption as cars get smaller and lighter, thus removing the incentive for automakers to downsize their lightest vehicles to comply. It also could mean that technology currently used to enhance horsepower would go instead to reduce gas consumption - a direct safety benefit because less powerful cars have lower crash rates.

Another way to conserve fuel, and serve safety at the same time, is to set lower speed limits. Going slower uses less fuel to cover the same distance. The national maximum 55 mph speed limit, enacted in 1974, saved thousands of barrels of fuel per day. It also saved thousands of lives. Highway deaths declined about 20 percent the first year, from 55,511 in 1973 to 46,402 in 1974. The National Research Council estimated that most of the reduction was due to the lower speed limit, and the rest was because of reduced travel. By 1983 the national maximum 55 mph limit still was saving 2,000 to 4,000 lives annually.

"Fifty-five was adopted to save fuel, but it turned out to be one of the most dramatic safety successes in motor vehicle history," Lund concludes. "The political will to reinstate it probably is lacking, but if policymakers want a win-win approach, lowering the speed limit is it. It saves fuel and lives at the same time."

I'm reporting this comment as:

Reported comments and users are reviewed by Autoblog staff 24 hours a day, seven days a week to determine whether they violate Community Guideline. Accounts are penalized for Community Guidelines violations and serious or repeated violations can lead to account termination.

    • 1 Second Ago
      • 6 Years Ago
      Screw it...helmets, safety cages, HANS, 5 point harnesses. How serious are you about safety???

      Maybe the best safety measure is using the gray matter between your ears and some training in accident avoidance.
        • 6 Years Ago
        Training doesn't do you any good if a pickup truck runs a red light at 50mph. And it also doesn't matter what you're driving in that situation, getting broadsided by an F-150 will kill you if you're driving pretty much any passenger vehicle.

        The only thing that would help in that situation is if the offending vehicle were smaller, which you have no control over.
      • 6 Years Ago
      And it took the government to tell us this. We all know it, they don't.

      As usual, they are the last to know. Can you imagine the NEW REGULATIONS they will come up with because of this NEW INFORMATION.

      God help us
        • 6 Years Ago

        No kidding!!!!!!!!!!!!!!!!!!!!!!!

        They are all of what you said they are. BUT we still did not need a test to show what we all already know.
        • 6 Years Ago
        It's the IIHS - not the government. "The Insurance Institute for Highway Safety is an independent, nonprofit, scientific, and educational organization dedicated to reducing the losses — deaths, injuries, and property damage — from crashes on the nation's highways." (from their website)
      • 6 Years Ago
      Isn't it curious that IIHS comes up with small car bashing (literally) and surprisingly revised tests during the times when small car manufacturers (overseas) are gaining and the Detroit, depending on large cars, is going south?

      Another thing - Europeans have been driving small cars for ages yet they have much better safety record than the US. We're talking about cca 37000 deaths in the US and about 33000 deaths in the EU that has much bigger population and much higher population density. The EU don't have that many pickup trucks, but do have incredible number of vans and trucks.
      • 6 Years Ago
      A charging elephant (SmartTwo or Yaris) against a wall at 40 mph is pretty impressive. Hits it, and not too much damage and structure seems intact. A blue whale (Hummer) against that small wall and it's even more impressive.

      Let the two meet each other somehow fac-to-face but at 60 mph and the impressive elephant becomes not so impressive.

      "Micro" vehicles only work in Europe and Asia because they are a country filled with micro vehicles. Small car hits small car = small accident. In the U.S. land of the Ford trucks, 90mph momma minivans, small Corolla is a big European sedan, when you drive a micro vehicle, you're just asking to one day "NOT LIVE" to appreciate your love of the planet.

      In an ideal world where we should all drive small vehicles, yes, the SmartTwo I'm sure is a very safe vehicle... but when you go into a gun fight, you don't bring a slingshot and a spitball pen.
      • 6 Years Ago
      Doesn't this show how barrier testing might not be a good thing? The differences between using a barrier and a real car are pretty shocking.

      Plus - didn't Mercedes do this with the Smart (with an E-Class) when it was new and it faired pretty good?
        • 6 Years Ago
        Not really, there's no way any one test would reflect all crash scenarios but a standardized barrier reflects how the vehicle itself manages collision energy and how that's transferred to occupants. If they were to make the crash test against another car, say a Honda Accord's front end. Then it would be possible to "cheat". You could design a vehicle that can absorb impact really well in the exact spots where a Honda Accord's hood and bumper would hit and get a five star crash test but the moment it hits something other than an Accord everyone's dead. Since we can't run every car into each other from every direction we're pretty well off seeing what happens if you hit a wall at 45mph.
        • 6 Years Ago
        Barrier tests are appropriate for determining what happens when you hit a fixed object (see telephone pole, guard rail, stopped car in front of you, etc...). There are still a lot of single-vehicle accidents. The test is also appropriate for determining what happens when you hit a similarly sized vehicle. So it's not a useless measure, just different.
        • 6 Years Ago
        That was a marketing trick. Mercedes sucks balls.
      • 6 Years Ago
      And yet the motor vehicle death rate in Germany at 3.1 per billion KMs traveled, where they have on average smaller cars than we do, is much lower than ours at 5.0 per billion KMs traveled.

      Source: Automobile Magazine
        • 6 Years Ago
        Largely, it's thanks to a serious problem of people in the US being misled into thinking larger cars are safer; ie: this pathetic "study".
      • 6 Years Ago
      Considering the general awfulness of damage that comes from head-on collisions, why don't the pictures show the larger car for comparison? For a head-on collision with a front impact like this, they also are probably pretty bad. Only showing their good side is just biasing things against the small cars.

      I can't find in the article exactly what the speed was, but it seems to suggest a 40MPG speed for each vehicle. 80MPH total head on collisions are very rare and will almost always work out in luck more than anything else. I'll bet if people saw two of the mid-size sedans hit each other at those speeds, they'd be aghast at the damage each would sustain.
      • 6 Years Ago
      Oh my god, you guys need to go on the IIHS website and watch the video pertaining to this test. At the end, they show a real time clip of the Smart crashing into the C-Class, and it really just looks like the C-Class punted it. Lol.

      • 6 Years Ago
      its funny, when they did that one test where they smashed a smart car into an S-class and into Cement Barriers, the host of that show (dont remember what is was) Praised it for its safety. This looks a HELL of a lot worse.
      • 6 Years Ago
      This test is lacking heavily on useful facts.

      First it's not comparable to the barrier tests as they are conducted with a single vehicle traveling at 40mph, where in this test both vehicles where at that speed and thus the resulting impact is at twice the barrier test speed. I would imagine that if you did the barrier test at 80mph you would see a similar result.

      It also assumes that if you crashed two mid-sized cars together they would both maintain their higher rating seen in this test. That seems like a stretch as the reason they probably scored better was the fact that their mass allowed them to resist the much smaller mass of the sub-compacts. Crash a mid-sized car in to a mid-sized car and I imagine things get mush less nice.

      Its a fantastic test for telling us that big things take less damage than small things when you crash them together, and as many have stated, no duh. It's also a great piece of propaganda for both the auto industry and the government to use for their own benefit.

        • 6 Years Ago
        Sorry, I'm clearly not an expert in this field and I'm not remembering my physics lessons perfectly. I wasn't factoring everything in and kind of made a big jump. If I had thought about it a bit more I would have remembered that things move exponentially in these scenarios and although it wouldn't be the equivalent of doubling the speed in the offset barrier test it clearly also isn't as Frank stated. It's more as Sean stated.

        I found this interesting:


        A 10 year old paper describing this exact phenomenon. Interestingly it points out what my gut was telling me that these tests are not equivalent.

        The Honda R&D test showed that a 50% offset at 56km in to a barrier had less severe results than two equivalent vehicles collided with the same offset and speed. Instead the vehicle-to-vehicle 50%/56km test results look more like a 40% offset crash at a higher speed (64km). This was with 1500kg weight vehicles.

        The paper addresses exactly what this recent test was about also.

        The NHTSA is clearly making things seem a bit worse than they actually are. Is it surprising that if you increase the crash severity small cars start to deform more aggressively than larger ones? Nope. Is demonstrating a scenario that is uncommonly more severe than an average crash to make a case against smaller cars really good science? I don't feel it is.

        Another interesting part of that paper is right in the beginning where it shows that most automotive deaths in the period they used (1991-1995) where at speeds well below the current speed limit. Maybe that has changed though. I haven't researched it fully myself.

        • 6 Years Ago
        "...where in this test both vehicles where at that speed and thus the resulting impact is at twice the barrier test speed. I would imagine that if you did the barrier test at 80mph you would see a similar result."

        Not correct. The result of hitting a barrier (that does not give) at 40 mph is the same as hitting a car of the same weight at that same speed (40 mph). A barrier crash at 80 mph would simulate two identical cars hitting with each of them going 80.
      • 6 Years Ago
      Bigger is often better.
      • 6 Years Ago
      Here's a radical idea:

      LEARN TO DRIVE.....the best way to avoid injury is to not wreck (duh). Pay attention to your surroundings . Get off the phone, stop texting, reading magazines and reviewing that speech for work. Always assume the other guy is going to do something stupid...that way when he/she runs the red light, it doesn't surprise you.

      I could never own one of those anyways, because I enjoy driving....and there is nothing fun about driving an econobox subcompact.
    • Load More Comments
    Share This Photo X