By on September 3, 2019

Honda Research family of dummies

Overwhelming geek that I am, I’m often reminded of The Simpsons in odd situations, mostly as the show’s been on so long, it can’t help but have covered a circumstance ad nauseum. In the car-centric season four episode “Mr. Plow,” the family heads to an auto show, where a suspiciously named automaker shows a slow-motion video of a crash test while Lisa watches. Disconcertingly, one of the dummies starts to crawl away from the scene of the crash, at which point the OEM rep shuts down the exhibit.

I did recently get the chance to watch a controlled car crash at Honda’s research facility. Just as important as seeing how the car handles the stresses of the impact is measuring how a human occupant reacts. Honda has a massive family of dummies, all ready to sacrifice themselves for the real people of the world.

Disclaimer: Honda brought journalists to the Columbus area and fed us two meals. As your author lives in Columbus, no hotel stay was needed.

No matter what The Simpsons would like you to think, automakers don’t use live humans in crash testing. How, then, can we be certain that the dummies indeed react like bags of flesh and bone? Honda turns to a climate-controlled lab deep within the bowels of the Research and Development center to constantly test the dummies for biofidelity, ensuring each of the scores of sensors strategically placed within each will accurately record impact.

Honda Research baby dummies

It’s not just the average-sized male, either. While the typical dummy used for frontal impact testing is the AM50 – adult male, fiftieth percentile size – there are options scaling up or down from there to ensure all possible occupants are protected. From infant and child dummies of 12 months, three years, six years, and ten years, to adult dummies AF5 (adult female, fifth percentile) and AM95 (male, 95th – basically your author with less facial hair), these dummies have seen it all. Further, specialized dummies for side impact or rear whiplash are available in the lab.

Honda Research THOR

This dummy looks a bit different than the rest, and it has a name – THOR. Sadly not named after the god of thunder, THOR is short for Test Human of Occupant Restraint. And rather than wielding a hammer, THOR is equipped with more than double the number of data channels (130 to 64) found on the customary Hybrid III dummy typically used. More data channels equate to more data and greater biofidelity. THOR, unsurprisingly, isn’t cheap – somewhere around $750,000 US. Thus far, however, THOR isn’t the industry or regulatory standard for crash testing, so the well-known Hybrid III remains the dominant dummy in the lab.

Brian Bautsch, manager and principal engineer for crash safety at Honda R&D, has a degree in biomedical engineering – in other words, the ideal education to become a dummy doctor. Here, Bautsch and his team sling a massive weight into the chest of a dummy to ensure that it is calibrated to react accurately to an impact.

Honda Research dummy test rig Honda Research dummy test 1 Honda Research dummy test 2 Honda Research dummy test 3

Adam Mihm, senior engineer for pedestrian protection, notes that 18 percent of all deaths on American roadways (6760 total) in 2017 occurred outside the vehicle – either pedestrians or cyclist. No matter the reason for the crash, automakers are working to minimize the damage caused to those not surrounded by two tons of steel. Here, by firing a model of a leg at the bumper of a Passport at 40 km/h (25 mph), engineers can measure the bending of the tibia and the elongation of four ligaments in the knee. Apologies for the reflection – Honda didn’t trust us barely-sentient dummies to not get hurt, so all of the tests were behind glass.

Honda Research leg impact 1 Honda Research leg impact 2

More weird reflections here, where a 2.5 kg (5.5 lb) ball sized like a child’s head is fired at the hood. Ever notice how most modern cars have a much taller hood than in decades past? Pedestrian impact protection is the reason, as space is needed between the hood and anything hard beneath – like the engine – to dissipate the energy of impact.

Honda Research head impact 1 Honda Research head impact 2

At the end of our day at Honda R&D, the masochists in the crowd were treated to a display that has to pain the hard-working engineers.

Two hundred and twenty-five times a year – roughly once each working day – a car is flung toward a 90 metric ton impact barrier that can rotate 360 degrees to measure any possible collision. Today, a 2019 Civic coupe (in the best color, Tonic Yellow) is hurtled down a long tunnel at 64 km/h (40 mph) to test small overlap, front driver-side crash performance.

Honda Research impact block

The impact area is lined with cameras – including underneath – and 13.6 million lumens of light to make sure the high-speed cameras capture everything.

Honda Research crash tunnel

No engineers or overfed journalists were harmed. Again, we were behind glass.

Honda Research crash 1 Honda Research crash 2 Honda Research crash 3 Honda Research crash 4

In this 64 km/h impact, that 90 ton barrier moves less than a millimeter.

Honda Research aftermath

After the crash, engineers swept the area of any shards of metal or glass, and we were permitted to come closer to see how the energy of the frontal overlap impact was spread throughout the car. The side-impact airbags were cut and tucked away so we could see the dummy – visibly shaken, but apparently unharmed.

Honda Research dummy aftermath

That’s how we all were, really – visibly shaken by the violence of the impact. It’s a daily occurrence for engineers at Honda and across the industry, of course, working to keep us dummies alive.

Honda Research dummy after crash

[Images: © 2019 Chris Tonn, crash video courtesy Honda]

Get the latest TTAC e-Newsletter!

9 Comments on “Safety for Dummies – Honda Braces for an Impact You Hope Never Comes...”

  • avatar

    Before anyone opines that older cars were safer, Here’s a video showing an offset head on collision between a 2009 Chevy Malibu and and 1959 Chevy Bel Air:

    We’ve come a long way in 50 (now 60) years.

    They sure don’t build ’em like they used to, and the modern car will mostly protect you circumstances that would have killed your (grand)father.

    Yeah, maybe the 1957 Bel Air would have fared a little better in some fender benders. However, the engineering ethics of building a car that totals itself leaving the occupants uninjured are about as ironclad of an engineering-ethics argument you could possibly make.

    They sure don’t build them like they used to, and that’s a good thing!

    • 0 avatar

      Oh lord the ’59 Bel Air video again. I think we should take a second to ponder how the GM X-Frame cars are the least-safe vehicles ever built.

      Of course no ’50’s car can hold a candle to modern crash safety, but when the only thing protecting you from the car T-boning you is the door skin, there have been some lapses in judgement.

  • avatar

    Do the dummies react to shards of Takata airbags going through them? Or do they need to test older vehicles for that?

    Before you post, “it impacts everyone,” Honda was complicit with Takata and their engineers doctored data to get airbags approved. Do a search – I’m not your research assistant. The finding was what Honda did was horrible and scummy, but of course, not to the level of criminal.

    • 0 avatar
      formula m

      You sound butt hurt… Did a GM heated seat melt the foam padding of the bottom seat cushion into a cone shape that was forcibly lodged into your ApaGtth upon a front end collision?

      • 0 avatar

        @Formula M: APaGttH is not wrong…

        Sounds like you’re butthurt that someone proved the lack of the integrity of Honda’s engineering staff. It sucks when you find out they’re just like everyone else…

      • 0 avatar

        Oh the irony…

        Recently I replaced the drivers-side seat heater pad in a Honda that I’m ready to sell because the old one shorted and almost burned through the leather. Yes, burnt foam padding underneath!

        Because I had to disconnect and remove the entire seat, I spent an hour last night learning how to reset the SRS system (airbags) so the indicator light would turn off. Thank god for YouTube… and airbags.

  • avatar
    Kyree S. Williams

    I can’t see myself getting a Civic Coupe. It just doesn’t communicate the right things, for me. But if I *were* to get a Civic Coupe, I would totally choose Tonic Yellow. It looks like it held up impressively in the small overlap, too.


    Volvo, for the past ten years or so, has been engineering its cars so that in a small-overlap collision, they sheer off the wheel on that side, allowing the rest of car to slide past the barrier while the sacrificed wheel dissipates crash energy. Other cars push the wheel rearward, toward the cabin, which “hooks” the car on the object of impact and causes that sudden deceleration that’s so injurious. Why is no one else able or willing to replicate Volvo’s strategy?

  • avatar

    Slightly off topic, but I saw a show on TV this weekend where a group was trying to design air bags for skiers (sort of a vest). The problem was that they would have to inflate before the fall, and how could the airbag tell if the skier was just doing a flip in the air or really falling?

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • Russycle: Yeah, my Element had that drivetrain, in a sedan it would be a hoot.
  • Russycle: Agreed. I like the Maxima’s looks the best, but that CVT. Can’t go wrong with the Honda.
  • jdmcomp: Volvo, Mercedes, VW, Audi, Subaru beg to disagree.
  • brn: I have not. I’m being narrow minded, based on my limited experience. :)
  • ajla: ” the best selling engine in the big mopars was the 318.” You have any stats to back that up?...

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Mark Baruth
  • Ronnie Schreiber