NHTSA’s (Not/Can’t Be) On It: Hyundai-Kia Vehicles With Substandard Theft Protection Continue to Wreak Havoc

S. Kane, E. Liberman, N. Black

The Hyundai Kia theft saga greeted the new year with a bang. Among other things: deaths in crashes involving stolen vehicles, an alleged kidnapping involving a theft attempt, a new municipal lawsuit, and another bright idea from Hyundai and Kia to protect vehicles which could not be fixed with a software upgrade – that will, no doubt be executed flawlessly. In other words, everything’s still totally bonkers. And yet, NHTSA, the nation’s esteemed regulator and enforcer of automotive safety, continues to observe from the sidelines.

In the first six weeks of 2024:

There have been at least seven deaths in crashes involving stolen Hyundai and Kia vehicles with teenage drivers at the wheel. According to a story in the Indianapolis Star, on Jan. 2, a teen driver slammed into another vehicle as he fled police in a stolen Kia Sedona. The driver of the other vehicle, 34-year-old Julio Cervantes Ramirez, died at the scene. Two days later, a stolen 2013 Hyundai Accent occupied by four teens crashed in Boston, killing a 14- and 15-year-old. Six days later, a third teen died from his injuries in that rollover crash. On January 14, another 14-year-old child died when the driver crashed a stolen Kia Sportage into a private residence; two other occupants were taken to the hospital. In mid-February, Michigan saw two more deaths in separate crashes involving stolen Kia vehicles.

A woman walking her dog on the streets of Northeast Washington D.C. had to scoop up her pet and take cover to avoid being hit by the driver of a stolen Hyundai Tucson barreling down the sidewalk. A couple in Columbus Ohio were charged with kidnapping after surprising a 13-year-old boy presumably trying to break into the woman’s Kia, allegedly forcing him at gunpoint into their apartment, and demanding cash from the boy’s family to fix the damaged ignition column. In a news report, Columbus City Attorney Zach Klein cautioned residents against administering their own rough justice, but expressed some sympathy for the defendants: “I do think it is evidence and a reflection of the frustration people have. They want to be able to maintain their livelihood and they know that may not happen if their car is stolen.’”

Newark, N.J. became the latest city to direct its ire at the manufacturers, suing Hyundai/Kia in U.S. District Court. Newark is trying to recoup more than $1 million the city spent in auto theft suppression overtime hours to address a car theft surge in the first ten months of 2023. The suit alleges that the amount is “more than the yearly auto theft suppression overtime costs of 2021 and 2022 combined.”  Like other municipalities, Newark saw a huge increase in auto thefts, with Hyundai and Kia models making up a disproportionate share of the vehicles reported stolen in the first 10 months of 2023. The lawsuit alleged a more than 1000 percent increase over 2022 theft reports, with nearly twenty percent of all of Newark’s registered Hyundais and Kias reported stolen in 2023, accounting for 58 percent of all vehicles stolen in Newark. 

The City of Austin joined the chorus of frustrated municipalities, passing a resolution urging NHTSA to compel a recall of Kia and Hyundai models without engine immobilizers. The basis for the resolution is auto theft figures showing that from November 2022 to November 2023 auto thefts increased by 63 percent; Hyundai and Kia vehicles accounted for more than a third of those thefts, even though fewer than 10 percent of vehicles registered in Austin are Hyundai or Kia vehicles. Said one councilman:

“Our hope is because we have multiple cities across multiple jurisdictions that our federal government will take notice and [the NHSTA] will see the importance of taking action to demand that Kia and Hyundai immediately recall their technology.”

One can dream. NHTSA is now, and has been, a rudderless agency, with blank spots on the organizational chart for Administrator and Chief Counsel. Increasingly, it’s hard to see how an overworked agency with no leadership at the top could take the sort of decisive action needed.

In a statement to KXAN in January, the Austin news station that sought a comment from the regulator, NHTSA assured the public that everything was under control:

This particular matter involves intentional criminal conduct under the purview of law enforcement authorities. However, since 2022, NHTSA has repeatedly met with Hyundai and Kia to discuss the causes contributing to the theft vulnerability, review the scope of differing software and hardware in the affected models, and receive regular updates on the companies’ action plans. NHTSA will continue to monitor this issue, spread awareness of further updates to local authorities, and lend its expertise in efforts to strengthen motor vehicle safety.

This statement ignores the entire history and intent of safety standard FMVSS 114 Theft Protection and Rollaway Prevention, which was promulgated in 1968 to address intentional criminal conduct under the purview of law enforcement authorities. NHTSA took up the issue because “casual” car thieves presented with easy targets were stealing vehicles and creating a safety hazard on public roadways.

Nonetheless, that’s more of a response from NHTSA than Safety Research & Strategies has gotten to its petition for rulemaking. Given NHTSA’s inability to enforce the requirements or address the intent of FMVSS 114 in the wake of the continuing waves of Hyundai/Kia thefts,   SRS made a formal request nine months ago that the agency amend the regulation. We noted that NHTSA has taken no compliance action at all claiming the theft protection safety standard was actually not enforceable as the problem continues to wreak havoc for owners, municipalities, law enforcement, and – in too many instances – causing carnage for pedestrians and other motorists who are being mowed over by kids fueled by exciting videos showing how they can play Grand Theft Auto in real life.

That’s because the federal anti-theft standard doesn’t mandate the use of engine immobilizers, instead, it gives automakers the choice to use effective technology – and the compliance test procedure for NHTSA’s contract labs doesn’t specify actions that include breaking or removing parts of the vehicle in an attempt to start it and gain forward mobility to determine compliance. Instead, the test procedure describes processes that simply specify removing the key (electronic or physical) from the vehicle, followed by an attempt to start the vehicle without it. But, NHTSA’s compliance procedure, which doesn’t lay out specific theft techniques for obvious reasons, does specifically state that the agency can authorize its contractors to depart from the procedures if they are consistent with the standard. The Purpose and Application section of NHTSA’s Laboratory Test Procedure for FMVSS 114 Theft Protection and Rollaway Prevention states:

The OVSC [Office of Vehicle Safety and Compliance] test procedures include requirements that are general in scope to provide flexibility for contracted laboratories to perform compliance testing and are not intended to limit or restrain a contractor from developing or utilizing any testing techniques or equipment which will assist in procuring the required compliance test data.

And

In addition, the laboratory test procedures may be modified by the OVSC at any time without notice, and the COTR [Contracting Officer’s Technical Representative] may direct or authorize contractors to deviate from these procedures, as long as the tests are performed in a manner consistent with the standard itself and within the scope of the contract.

Back to our Petition for Rulemaking: By regulation, NHTSA is supposed to notify the petitioner within 120 days of its decision to grant or deny the petition. But, other than an acknowledgement that NHTSA received the petition, we’ve heard nothing.

Meanwhile, Hyundai and Kia’s ad hoc service campaigns continue, as do reports from owners that even after Hyundai and Kia software upgrades were done, their vehicles are still easy to steal quickly with little technical skill or special tools. The situation is unique to the U.S. It is not occurring in Canada or Europe, where immobilizers have been required since 2007 and 1998 respectively. So, elsewhere, the Korean automaker is fitting the same model vehicles with anti-theft engine immobilizers. But in the U.S., where its peers have long equipped their fleets with these effective features voluntarily, Hyundai-Kia decided it wasn’t necessary, because it could install any anti-theft measures it wanted. In America, Hyundai-Kia is an outlier – by choice – but wants its bad decisions to be viewed as if it’s all about everyone else’s actions.

In Omaha, a 2019 Hyundai Tucson was stolen the same way they are all stolen with a busted back window and a USB cord, despite receiving the software upgrade in August. According to a news report:

Chuck [Peters] purchased the Hyundai for his daughter and took it in after the August recall to install anti-theft software. The decals on the windows telling the world a starter inhibitor has been installed, the anti-theft technology would stop any attempt to steal the vehicle — but the car was stolen. The thieves left behind broken parts and an empty parking space. Police recovered the car, but when Chuck went to get some information from his daughter’s vehicle, he forgot to take the keys.

“What was really troubling for me was when we got the vehicle back, I had to go to the body shop to get some information for our insurance company. I was able to start the vehicle using the USB cable method as well, and that’s exactly what that update is supposed to fix,” he said.

In Louisville, 84-year-old Bobbie Sanders reported that her 2020 Kia Rio was stolen twice – once out the Kroger’s parking lot as she shopped, and a second time, just recently, out of her driveway. Each theft occurred after she took Rio to the dealer for a software upgrade in May and a second upgrade in October.

“We’ve done two upgrades with Kia and my mom still doesn’t have a car,” [her daughter Elizabeth] Madden said. “My mom’s out of all the monthly payments that she’s paid, the insurance that she’s paid, as well as the $1,000 loan she’s paying on that she had to pay to get her car out.”

In Milwaukee, Trisha Nguyen’s 2014 Kia Optima has been stolen three times – at least once after she had the software re-flash.

“I was initially told that the car could only start with a key ignition but that’s not the case because they broke the steering column again, and started it with a USB cord.” said Nguyen.  

In October, TMJ4 News spoke with a project manager for Kia’s Anti-Theft Program who explained how the upgrade is supposed to work.

‘This software update makes it so even if they try to plug that USB port and it’s not going to disable the ignition immobilizer, it’s going to keep that intact and it is going to sound the factory alarm,’ said Emily Falecki, Project Manager with Kia’s Anti-Theft Program.

On Friday, Falecki told TMJ4’s Ryan Jenkins by phone that part of this upgrade requires Kia owners to lock their cars with a key fob. If the vehicle isn’t locked with the key fob, the upgrade doesn’t work.”

And now, Hyundai and Kia have announced a new fix for vehicles that were ineligible for the software update: the installation of an “ignition cylinder protector with a locking bracket.” In a December 15 news release, Hyundai stated that protectors had been “independently tested and verified by a leading engineering and scientific consulting firm. It reinforces the ignition cylinder body and prevents its removal through the method of theft promoted across social media.” Kia followed with a similar announcement on December 29.

The repair procedure involves the installation of a metal shield that’s installed under the plastic steering column cover that’s easily broken and allows access to the ignition cylinder. To install the shield, the plastic steering cover is removed, then the key portion of the ignition cylinder is removed (along with the illuminated key ring if equipped). The ignition cylinder is then reinstalled after it’s coated with epoxy that permanently bonds it into the ignition assembly, and a metal protective shield is inserted over the ignition cylinder with more epoxy and further secured into place with screws that have break-away heads. Once this is complete, anti-theft decals are added to the vehicle side windows. (Vehicles with illuminated key rings will no longer have that lighting function restored.)

The Kia service procedure generally mirrors the Hyundai procedure described, but Kia specifies a different brand of epoxy (Loctite versus J-B Weld). And Kia requires the driver to sign a waiver noting that if the customer vehicle has a key illumination ring, it will be permanently removed, any future replacement of the ignition switch will require complete replacement of the ignition cylinder assembly, and that reinstallation of a new theft deterrent ignition cylinder protector that is compliant with the service program “will be at no cost to you.” The language “any reinstallation of a new Theft Deterrent Ignition Cylinder Protector will be at no cost to you” appears to indicate that in the event of a second installation there will be no cost to Kia America, because the rest of the waiver addresses the signer, (i.e., the customer) as “I.”

The repair process requires technicians to use epoxies that are extremely sticky and can be quite difficult to contain when the parts are put together, potentially causing excess epoxy to contaminate the electrical and mechanical components – an issue acknowledged by Hyundai and Kia in their technical bulletins. And all of this is happening when technicians are working under time constraints of labor allotments ranging from 15 to 40 minutes for the job, depending on the model.  

The completely permanent bonding and securement of the ignition component parts will require entire ignition switch assembly replacement in the future if the lock cylinder or other related components are damaged – either from theft, attempted theft, or other failures, including wear.

Hyundai plans to install the device in 646,000 vehicles, including the 2011-2017 Accent; 2013-2014 Elantra Coupe; and the 2011-2012 Elantra Touring, Genesis Coupe, Santa Fe and Veracruz vehicles. Kia’s service campaign included Soul, Rio, Forte, Sedona, and Sportage models from various production date ranges, so it is hard to tell exactly which model years are affected.

Hyundai’s press release indicated that starting on Dec. 20, it planned to reach out to owners about the new anti-theft service campaign via mail, email, phone, social media outreach, search engine marketing and display marketing. Both automakers directed owners to use VIN look-up tools at their dedicated websites to receive instructions on how to get the repair. We tested VINs for eligible Hyundai models this week, and Hyundai’s anti-theft customer page still provides no information. (But a call to Hyundai’s customer service with specific VINs verifies eligibility for the shield.) 

Across the country, large cities are posting high triple-digit increases in their theft rates, with thefts of inadequately protected Hyundai-Kia vehicles taking up a disproportionate share. For example, in August, Chicago sued Hyundai-Kia, alleging: “In 2022, more than 8,800 Kia and Hyundai vehicles were stolen in Chicago alone. This figure represented 41% of Chicago’s car thefts, even though Kia and Hyundai vehicles made up just 7% of the vehicles. Unfortunately, that trend has continued into 2023 and does not appear to be slowing.”

In December, the Highway Loss Data Institute (HLDI), which tracks insurance claims, released its third analysis in two years regarding theft claims of Hyundai-Kia vehicles without immobilizers. It showed a 1000-percent increase in such claims from the first half of 2020 to the first half of 2023. And the rise was far above its peers: in the first half of 2020, Hyundai Kia vehicles, like those of other manufacturers, were reported stolen at a rate of 1 per 1,000. In the first half of 2023, Hyundai-Kia theft reports had skyrocketed to 11.2 per 1,000, while the rate for other vehicles remained flat.

Uncaptured by theft figures are what seem to be significant numbers of crashes, deaths, and injuries involving stolen Hyundais and Kias. While no official entity appears to be gathering this data, these incidents crop up in the news on an alarmingly regular basis.

But, don’t worry folks, NHTSA is totally on it.

The Hyundai Kia Theft Mayhem Continues; SRS Keeps NHTSA Apprised

On April 27, Safety Research & Strategies petitioned the National Highway Traffic Safety Administration for rulemaking to revise the compliance test for FMVSS 114, Theft Protection and Rollaway Prevention. The request was prompted by the rising crashes, deaths and injuries linked to Hyundai/Kia vehicles with inadequate theft protection, and by NHTSA’s reluctance to take any enforcement action.

The tsunami of thefts began in 2021, when car thieves in Milwaukee learned to take advantage of a vulnerability in MY 2010-2021 Hyundai and Kia vehicles with traditional metal keys. The numbers began to jump that July when a video posted to TikTok demonstrated how to exploit the lack of meaningful theft-prevention features in Hyundai/Kia vehicles by evading the rudimentary burglar alarm, removing the plastic steering column shroud to access the ignition cylinder and with household items and no technical skill, start the engine and drive off. The how-to video quickly went viral, with mainly teenaged thieves posting their exploits under the hashtag “Kia Boys.”

At the time, The Safety Record provided its readers with the regulatory history of FMVSS 114, from the standard’s origins and intent to the changes that were meant to address the evolution of automotive technology, but have rendered the standard so ineffective, that it no longer acts as a safeguard against theft or rollaway.  

Today, SRS amended its petition with updated crash and harm numbers, to stress the importance of revising the current compliance test from one that is nearly impossible to fail, to one that reasonably assesses whether the system in place deters casual thieves. The data also support the need for an effective recall, not simply a “customer satisfaction” campaign, because it’s increasingly clear these vehicles contain a defect that represents an unreasonable risk to motor vehicle safety.      

Citing the limitations of FMVSS 114, NHTSA has continued to resist calls from state and municipal government officials to force Hyundai and Kia to recall nearly 9 million 2011 to 2022 models with inadequate theft prevention features for failing to comply with the standard.

(Korean-based manufacturers Kia Corp. and Hyundai Motor Corp. are affiliates of Hyundai Motor Group. Both companies are involved in the joint design and development of vehicles sold globally under their respective brands. Hyundai Motor Co. owns 33.88 percent of Kia Corp., making it the largest investor in the company.)  

Hyundai and Kia have instead issued customer satisfaction campaigns that include updated software for some affected models and aftermarket steering wheel locks for others, in attempt to quell the theft epidemic that’s largely been driven by kids. The new software updates the theft alarm logic to extend the alarm sound from 30 seconds to one minute, and requires the key to be in the ignition switch to turn the vehicle on. With the new software, locking the doors with the key fob sets the factory alarm and activates an “ignition kill” feature which customers have to use the key fob to unlock their vehicles to deactivate. Hyundai/Kia also offered window stickers announcing the presence of anti-theft software. Owners of vehicles ineligible for the software fix got a reimbursement for a steering wheel lock.

Previously, in vehicles equipped with a burglar alarm, the system would only prevent an engine start if the burglar alarm was activated – a major design oversight. These updates disallow an engine start when bypassing the ignition lock cylinder, and rotating the ignition switch, but they only work when the burglar alarm is armed, meaning that the doors are locked using the key fob or the key in the driver’s door. However, if the alarm was never triggered (because entry was gained via the broken window – the way many of these thieves have operated), the vehicle could still be started, and the ignition kill feature would not be activated, even though the alarm was still armed.

Perhaps this is a reason why those “fixes” have not done nearly enough to halt the utter chaos caused by the combination of a TikTok video, showing budding car thieves just how quick and easy is it to steal a Hyundai or a Kia, with a minute and a USB cable, and two automakers who eschewed immobilizers. According to the Highway Loss Data Institute, while 96 percent of automakers had immobilizers as standard equipment in MY 2015 vehicles, only 26 percent of Hyundai/Kias were so equipped.

During the last two years, the vulnerabilities of Hyundai/Kia vehicles without immobilizers have caused an astonishing number of thefts, crashes, injuries and deaths, along with other violent crimes – carjackings, homicides, and burglaries (ramming a car into a storefront to break in), to name a few, committed in a stolen Hyundai or Kia. Some of these incidents involve drivers who are too young to get a driver’s license. For example, in July 2022, two 14-year-olds in Columbus, Ohio died, and a third was injured, when they crashed a stolen Sonata into a warehouse, ejecting two occupants and trapping a third inside the vehicle. A year later in Orlando, Florida, a 15-year-old driver ran a red light at high speed, striking an SUV and killing a 23-year-old man. The stolen Santa Fe had five passengers: ages 16, 15, 14, and 13 years.

“Unfortunately, this incident – with a very young, inexperienced driver crashing a stolen Hyundai/Kia vehicle, causing deaths and injuries – has become all too common,” says Attorney Frank Melton of the Florida firm Newsome Melton, who is preparing to file a case in the Orlando incident. “The automaker has a responsibility to ensure that kids who have no technical training and aren’t even eligible for drivers licenses can’t breach the vehicle’s anti-theft features in less than two minutes – and NHTSA should use its statutory enforcement tools to hold automakers to that obligation. In the absence of either, accountability will move to the courts, as the damage continues.”

In May, a 12-year-old and 13-year-old in Hamden, Connecticut were arrested after crashing two different stolen vehicles.

The total number of crashes, injuries and deaths, is, as yet, unknown. No single entity appears to be officially tracking them; news reports are the current source. On February 14, when NHTSA announced the launch of the campaign, it linked these thefts to at least 14 reported crashes and eight fatalities. That was already an undercount. By then, the media had written about 42 crashes, 27 injuries, and 21 deaths, from June 2021, when a 16-year-old boy from Milwaukee in a stolen Kia Sportage died after a police chase and head-on crash with an SUV, which left five occupants seriously injured, to February 12, 2023, when three 13-year-old boys were arrested after allegedly stealing a Kia and crashing into another car, killing a 71-year-old man.

Since then, the news media reported another 90 such crashes, resulting in 23 more deaths (including a six-month-old boy, a four-year-old boy, and a 14-year-old driver), 99 injuries, some of which were said to be serious injuries, and one house fire. In addition, these thefts have caused structure damage and other vehicle damage, including seven police cruisers, a fire engine and a school bus, caused by, often youthful, drivers of stolen Hyundai/Kia vehicles. In total, using new stories as the sole source, from June 2021 through October 12, we identified 132 crashes, 44 deaths and 126 injuries. Again, this is likely an incomplete accounting.

The theft rates have been through the roof – in many cities disproportionately higher in the Hyundai/Kias that lack immobilizers than in any other competitors’ models. In April, California Attorney General Rob Bonta and 16 of his counterparts across the nation sent a letter to Acting NHTSA Administrator Anne Carlson asking NHTSA to compel a recall of MY 2011-2022 Hyundai and Kia vehicles without immobilizers. The letter stated:

For example, in Los Angeles, thefts of Hyundais and Kias increased by approximately 85% in 2022, and made up almost three quarters of the entire increase in stolen cars of any make and model in the city. 7 Hyundais and Kias also constituted approximately 20% of stolen cars in Los Angeles in 2022, up from 13% in 2021. Similarly, in Berkeley, California, thefts of these cars have made up 38% of vehicle thefts since the end of 2022. California cities’ data is consistent with data from other states. For instance, in Minneapolis and St. Paul, Minnesota, thefts of Hyundais and Kias increased 836% and 611%, respectively, in 2022. In Columbus, Ohio, Hyundais and Kias constituted nearly 45% of stolen cars in 2022, in Milwaukee, 58%, and in Minneapolis, 33%.

Data from a July civil complaint filed by the 17 cities in seven states against Hyundai/Kia in a California federal court, contains a city-by-city account of the precipitous and continuing rise of thefts, crashes, injuries, death and crimes associated with Hyundai/Kia thefts. For example, Madison Wisconsin reported that between 2021 and 2022, thefts of Kia vehicles rose by 124 percent; in the summer of 2022, thefts of Kia and Hyundai automobiles increased by 270 percent, accounting for more than half of the auto thefts there.

Atty. General Bonta noted that the Hyundai Kia vehicles with inadequate theft protection violated the requirements of FMVSS 114:

Specifically, FMVSS Number 114, S5.1 requires vehicles to have “a starting system which, whenever the key is removed from the starting system prevents: (a) [t]he normal activation of the vehicle’s engine or motor; and (b) [e]ither steering, or forward self-mobility, of the vehicle, or both.”

The rampant theft of Hyundai and Kia vehicles makes clear that these vehicles’ starting systems do not prevent engine activation, steering, or forward self-mobility when the key is removed from the starting system. Indeed, because the vehicles have easily bypassed ignition switches, a screwdriver and USB cable are sufficient to start and drive off with the cars in a matter of seconds or minutes—no key required. Such starting systems do not meet FMVSS Number 114’s requirements. The lack of engine immobilizers in these vehicles, which could have provided a second line of defense against theft, has compounded and exacerbated this problem.

Additionally, these Hyundai and Kia vehicles’ vulnerability to theft constitutes a defect posing an unreasonable risk to safety, providing NHTSA with an independent basis to order a recall. Even young teenagers are able to access the ignition system and drive off in these vehicles

And, just as The Safety Record pointed out right after the fix was announced, Atty General Bonta noted that a customer satisfaction program allows the automakers to bypass the notice and accountability requirements of a formal recall. The Attorneys General complained that such a program would roll out too slowly and reach too few.

In a June letter, Cem Hatipoglu, NHTSA’s acting associate director for enforcement, responded to Bonta and the other AGs, saying that NHTSA wasn’t inclined to take any action other than monitoring the situation. He wrote:

At this time, NHTSA has not determined that this issue constitutes either a safety defect or noncompliance requiring a recall under the National Traffic and Motor Vehicle Safety Act, 49 U.S.C. Chapter 301. The Federal Motor Vehicle Safety Standard identified in your letter, FMVSS No. 114, does not require an engine immobilizer. See 49 C.F.R. § 571.114. Also, the test procedure specified in that standard does not contemplate actions taken by criminal actors to break open or remove part of the steering column and take out the ignition lock to start a vehicle. [Emphasis added.]See id. § 571.114, S6. Here, the safety risk arises from unsafe use of a motor vehicle by an unauthorized person after taking significant destructive actions to parts of the vehicle.

Hatipoglu’s assertion that the compliance test doesn’t address hotwiring is technically true, but NHTSA absolutely did consider that very reality in promulgating the standard, and in a 2004 interpretation letter from then-NHTSA Chief Counsel Jacqueline Glassman noted that in a response to an unidentified automaker requesting guidance on its engine immobilizer and the requirements of FMVSS 114. Glassman agreed that the system the automaker described would be compliant with FMVSS 114 because if an attempt was made to circumvent the ignition lock, including through “hot-wiring,” the immobilizer prevented engine starting without the key.

The current compliance test basically allows the tester to sit in the vehicle without the physical key or key fob, and try to start it. Pretty impossible to fail. Immobilizers are optional, but automakers that install them must pass specific and more rigorous test requirements, based on the standard in Canada, where immobilizers are required.

According to 49 CFR Part 552.8 .8 Notification of agency action on the petition, NHTSA was supposed to get back to us by the end of August:

After considering the technical review conducted under § 552.6, and taking into account appropriate factors, which may include, among others, allocation of agency resources, agency priorities and the likelihood of success in litigation which might arise from the order, the Administrator will grant or deny the petition. NHTSA will notify the petitioner of the decision to grant or deny the petition within 120 days after its receipt of the petition.

As vehicles become more complex, the need for strong NHTSA leadership and adequate resources becomes more urgent. And yet, the agency has long been stymied by the lack of a confirmed administrator with enough tenure to provide it. From April 2017 to May 2022, NHTSA has been run by a series of short-term acting administrators. The last administrator, Steven Cliff, was nominated by President Biden in January 2021, but not confirmed by the U.S. Senate until May 2022. He left in August 2022 for the California Air Resources Board. Biden nominated the current Acting Administrator Ann Carlson on February 13, but was forced to withdraw it in May, after Republicans opposed her green initiatives, such as rigorous fuel efficiency requirements.

It is difficult for a public health agency, with a safety mission as critical as NHTSA’s, to fulfill it while rudderless. The ongoing Hyundai-Kia debacle is but one result.

Hyundai/Kia Recall Vehicles with Leaking ABS Modules for the 17th Time in Seven Years

Late last month, Hyundai/Kia recalled nearly 3.4 million vehicles, warning owners to park their vehicles well away from structures – like a house or garage – because a leak in the ABS module can cause an electrical short and lead to a fire. Hyundai’s campaign involves 13 different MY 2011-2015 models for a total recall population of 1,642,551 vehicles. Kia is recalling more than 1.7 million vehicles encompassing 12 different models, covering model years 2010-2019.

(Korean-based manufacturers Kia Corp. and Hyundai Motor Corp. are part of Hyundai Motor Group. Both companies are involved in the joint design and development of vehicles sold globally under their respective brands. Hyundai Motor Co. owns 33.88 percent of Kia Corp., making it the largest investor in the company.)  

This recall is mega-sized, but it is nothing new. These two campaigns represent the 16th and 17th recalls from Hyundai and Kia for the same problem in seven years. They involve a common component by different names: in Hyundai vehicles, it’s called the Anti-Lock Brake System (ABS) module, while in Kia vehicles, it’s called the Hydraulic Electronic Control Unit (HECU). They have been manufactured by two different suppliers: Mobis and Mando. But 14 of the recalls, including all of those launched from 2020 to the present, involve components manufactured by the latter. They all share a common design configuration in that they are Powered All the Time (PAT), meaning even when the vehicle’s engine and ignition system is off, these modules still maintain a flow of electrical current. The recalls warn owners not to park their vehicles in or near a structure, because the engine or ignition doesn’t have to be on for a fire to start.

This is reminiscent of the Ford Cruise Control Deactivation Switch (CCDS) debacle, which eventually resulted in six recalls covering 14.9 million vehicles from 1996 to 2009. Ford, and later NHTSA, studied fires and thermal events involving this switch for eight years before it determined that a number of factors set off the chain of events that resulted in underhood fires: material fatigue, the orientation of the CCDS above the brake master cylinder, and the PAT configuration. The failure of the switch seals allowed for fluid intrusion, which could, eventually, lead to an electrical short and subsequent fire, including when the ignition and engine were off because the switch was always powered. Amid seven separate NHTSA investigations, Ford shifted the blame from defective switches damaged in the manufacturing process, to a systemic problem influenced by the switch’s position and the electrical architecture of the cruise control system, to an age degradation issue. The defect has been blamed for at least 1,500 fires – many in parking lots and garages – causing hundreds of thousands of dollars in property damage, and is alleged to have caused at least three injuries and three deaths.

Similarly, Hyundai and Kia have been “investigating” and recalling defective ABS/HECU modules since November 2016, with the vast majority of campaigns launched in the last three years. These latest recalls for leaking ABS modules brings the total recall population to nearly six million vehicles. Roughly half of the recalls name moisture or brake fluid intrusion into the modules as the likely root cause, while the other half state the root causes were undetermined, even though some suggest there are signs that contaminants are leaking into the HECU. When Hyundai and Kia assigned blame, it pointed to supplier quality, including improperly sealed wire harness covers, or excess flux residue from the soldering process at the supplier, compounded by exposure to heat/humidity and deteriorating seals. But many of the recalls involving Mando modules concede outright that despite a prolonged joint investigation, they were unable to determine the cause of the short circuits.

Hyundai’s most recent Mando ABS module recall pinpoints the problem as defective O-rings:

Certain ABS motor shaft O-ring material formulations may be susceptible to physical changes over time due to varying factors, including vehicle ABS specifications and/or the presence of foreign contaminants in the brake fluid, such as moisture, dirt, and dissolved metals, which could affect sealing strength and result in brake fluid leaking onto the ABS controller PCB.

Kia’s version of the recall said that it really wasn’t certain why it was recalling so many of its vehicles. Its Part 573 submission to NHTSA said: “It is believed that a short circuit may result in excessive current within the HECU [Hydraulic Electronic Control Unit]. Exact cause of the short circuit remains unknown.”

In its mandated Defect Chronology submission to NHTSA, Hyundai asserted that its American and Korean engineers have been investigating this problem for four years, beginning with analyses of some overheated ABS modules it collected in 2019. (Hyundai did not mention the earlier investigation that led to a recall in 2018.) Then, beginning in the summer of 2020, as Hyundai and Kia were initiating several recalls into other models and model years for ABS/HECU-related fires, the lawsuits related to fires in this new subset of vehicles began to roll in. That August, Hyundai’s North American Safety Office opened a formal case investigation beginning with Elantra vehicles, and began to recover all similar ABS modules with thermal damage from American market vehicles. Throughout most of 2021, Hyundai continued investigating. In some cases, they found brake fluid in the switch but could not determine how it got there. In October 2021, Hyundai brought in Exponent and after nearly two more years testing and analysis, identified the cause as the O-ring degradation within the ABS module, which lost strength and deformed over time breaking the seals:

During this period, Exponent focused testing on the motor shaft O-ring material durability. In an update provided on May 9, 2023, Exponent reported that foreign contaminants were present in residual brake fluid found inside certain analyzed ABS modules. In an update provided on July 7, 2023, Exponent found the O-rings used in the subject ABS modules consisted of varying rubber material formulations determined through thermogravimetric (“TGA”) analysis. In an update provided on September 6, 2023, Exponent confirmed that certain material formulations used in the O-ring rubber could lose hardness over time. Additionally, the material could be affected by foreign contaminants in the brake fluid, potentially impacting sealing performance. Based on this information, HMC conducted a review of supplier manufacturing records and confirmed a material formulation change to one with increased hardness implemented by the supplier in September 2014 and February 2015 at the Korea and U.S. plants, respectively.

Based on this revelation about a materials change, Hyundai recalled vehicles from the 2015 model year and earlier.

Kia starts its Defect Chronology in 2023, claiming that in July it learned that Hyundai was investigating overheating ABS modules (HECUs in the Kia vehicles), and noticed that those modules were in many of its models. During the ensuing two and a half months, it launched its own internal investigation and found some similar overheating incidents and some leaky modules but, once again, was unable to pinpoint a root cause. But given Hyundai’s go-big recall, Kia apparently decided to do the same. Its investigation is less likely to blame pre-2015 Mando O-rings, because its recall population also includes vehicles from the 2016 to 2019 model years.

You might be thinking that they would address the enduring problems with its ABS/HECU modules by simply replacing all of the older modules with newly designed modules.

You would be wrong. Despite this obvious hardware solution, Hyundai/Kia have only offered module replacements in four of the early recalls. In three other recalls, the fix was to install a relay in the vehicle’s main junction box to prevent the risk of an ABS short-circuit while the car is turned off. Of course, that wouldn’t prevent a thermal event when the engine was running. However, in most of the recalls, including all of those from the last three years, the only remedy is the installation of lower-amp fuses, so that if there is an electrical short, it will blow the fuse rather keeping the module from overheating and igniting.  While that may prevent fires, if the vehicle is in motion when this occurs, it could force the driver to stop operating the vehicle, which could create a new danger.

What all of these recalls really appear to demonstrate is that the only permanent remedy is to replace the ABS modules with a design that functions throughout the vehicle lifecycle without leaks into the electrical unit.

But Hyundai and Kia aren’t doing that.

Working the Refs

In just two months, Hyundai is expected to emerge from the shadow of a three-year Consent Order it signed with NHTSA just before Thanksgiving 2020. Back then, the agency slapped Hyundai/Kia with a combined $210 million civil penalty, the largest in the regulator’s history, for lying (more politely called “inaccuracies” and “omissions”) in its communications with the agency and failing to launch timely recalls involving more than 1.6 million Hyundai and Kia models with the company’s Theta II engines. The engines were supposed to make Hyundai/Kia the leaders of fuel economybut instead suffered from manufacturing and design defects that caused lubrication problems and premature bearing wear that could lead to loss of motive power and fires. (Read about it here: Hyundai-Kia’s Billion Dollar Engine Problem that Broke the NHTSA Civil Penalty Barrier.)

The story is instructive for its similar trajectory of multiple years of rolling recalls and satisfaction campaigns to address a problem the “fixes” didn’t correct. Hyundai alleged that the engine problem began in 2011, when its Montgomery, Alabama, assembly plant changed the way it removed machining debris from the crankshaft of the then-new Theta II Gasoline Direct Injection (GDI) engine. The warranty claims for excessive noise, an illuminated check engine light, and stalls began to rise as soon as the Theta II debuted in the field.

By 2015, the complaints accruing in NHTSA’s VOQ database began to concern the agency, which contacted the automaker, worried about the potential for high-speed stalls. And so in September, Hyundai recalled 470,000 Model Year 2011-2012 Sonata vehicles equipped with 2.4L and 2.0L Theta II GDI engines.

The defect involved the travel of metal debris generated during engine crankshaft manufacturing and left in the component’s oil passages, and travel, over time into the connecting rod oiling passages, restricting oil flow to the bearings,” which could raise engine temperatures and lead to premature wear of the connecting rod bearing, eventual failure, and a vehicle stall and a potential fire. The remedy was an engine noise inspection that required moving the vehicle to a quiet place and positioning a mobile tablet near the steering wheel to assess the engine sound, while an unexplained algorithm determined if the vehicle passed or failed the inspection. The failures got a new engine; the others got a new dip stick and an oil top-off.

During the following two years, Hyundai and Kia, which also had models with the Theta II engine, attempted to avoid a recall with extended warranties, even as Hyundai engineer and 26-year company veteran Kim Gwang-ho blew their cover. A member of Hyundai’s Quality Strategy team, Kim traveled to the U.S. in August 2016 to allege in a meeting with NHTSA that Hyundai’s recall did not cover the entire population of affected vehicles in the U.S. and South Korea, and that the problem was also related to the engine design. 

In March 2017, Hyundai and Kia finally announced recalls for a combined nearly 1.2 million vehicles describing the same defect as the 2015 recall. Two months later, NHTSA responded by opening a Recall Query to determine if Hyundai and Kia met its regulatory burden of issuing these recalls within five days upon learning of a defect.

On November 23, 2020 everyone signed the Consent Order agreeing that they hadn’t.

Under its terms, Hyundai and Kia had to spend a significant portion of the fine on internal process improvements. ($40 million and $16 million respectively.) Hyundai was required to “build and develop a fully functioning United States-based outdoor test laboratory and vehicle tear down facilities. The test laboratory will focus on safety field issues, vehicle inspections, and defect investigations.” Finally, Hyundai/Kia was required to hire a third-party auditor to oversee its progress. The penalty capped a five year period in which millions of Hyundai and Kia models were scrutinized under four different investigations, and were the subject of class-action lawsuits, extended warranty programs, a product improvement campaign, and 10 recalls to deal with defects that cause stalls and non-crash fires attributed to a variety of causes.

According to the NHTSA press release on the agreement, both companies were required to “develop and implement sophisticated data analytics programs to better detect safety-related concerns.” The Safety Record wonders what these new data analytics are telling them about leaking ABS modules.

And as Hyundai/Kia enters its seventh year of placing Band-Aids on ABS modules in the apparent hope that NHTSA doesn’t notice that their recall “remedies” may be as crappy as the modules themselves, may we remind you that Hyundai/Kia still has not launched a recall to fix their theft-prone, immobilizer-free models.

According to Hyundai/Kia’s defect chronologies, the defective ABS modules, have not caused any reported deaths or injuries. In contrast, the Hyundai/Kias with insufficient theft protection have touched off a two-year nationwide epidemic of criminality that shows no signs of abating, with steep car theft numbers and a horrific trail of crashes, injuries and deaths. The last two months alone have seen Hyundai/Kia theft news stories with headlines such as: “3 dead, 2 critical after stolen car ‘cut in half’ in crash with pole at MLK, Carey” (Las Vegas); “21-Year-Old Dead, Passenger in Critical Condition After Being Hit By Teens Driving Stolen Kia in Old Brooklyn” (Cleveland); or “12-Year-Old Seriously Injured After Smashing Stolen Kia Into a Utility Pole: Police” (Bridgeport).

This is just a small sample; there have been many more since February when NHTSA announced that, after eight deaths and 14 crashes attributed to stolen unprotected Hyundai and Kia vehicles, the Korean conglomerate would launch a service campaign to apply a software fix and a decal to vehicles without immobilizers. No apparent investigation, no recall.

In June, Cem Hatipoglu, NHTSA’s acting associate director for enforcement, responded to a demand by state Attorneys General that NHTSA exercise its authority to compel meaningful action, by saying that NHTSA wasn’t inclined to take any action other than monitoring the situation. He wrote: “At this time, NHTSA has not determined that this issue constitutes either a safety defect or noncompliance requiring a recall,” noting that the compliance test “does not contemplate actions taken by criminal actors to break open or remove part of the steering column and take out the ignition lock to start a vehicle.”

This is true even though FMVSS 114 was promulgated for this precise reason: to prevent casual thieves from quickly hot-wiring a vehicle for high-speed joy rides that end in tragedy. That’s why Safety Research & Strategies has petitioned NHTSA to strengthen the anti-theft compliance test provision. (You can read about it here: The Amazing Shrinking 114 )

 

The Amazing Shrinking 114

Federal Motor Vehicle Safety Standard (FMVSS) 114 Theft Protection and Rollaway Prevention began its life in 1969 as solely a theft protection standard in response to a wave of amateur car thieves and the attendant mayhem on the road. By the late 1980s, that rule began to tackle the problem of rollaways caused by the ability to inadvertently shift the gear selector into non-Park positions with the key removed from the ignition. In the mid-2000s, the agency began another round of rulemaking to address the advent of keyless ignitions. But the standard has not been amended in 17 years, and during that time it has drifted far from its historic intent, providing the public with little theft protection or rollaway prevention. Whether the issue is deadly rollaways in keyless ignition vehicles or a tsunami of auto thefts, NHTSA has found no recourse in FMVSS 114 – the standard it wrote to address these hazards. Be it rulemaking, compliance, or enforcement NHTSA has consistently reacted to issues that underscore the standard’s weakness by declining to use its authority to uphold the standard’s intent. That is especially true if identified safety hazards can be blamed on drivers, rather than designs which encourage human errors without adequate countermeasures.

In 2021, Congress broke a decade-long impasse to update the standard with the passage of the Infrastructure and Jobs Act. The law compels NHTSA to pass a Final Rule requiring vehicles with keyless ignitions to install an automatic engine idle shutdown to prevent carbon monoxide poisonings. Ford and GM have been doing this beginning in 2012. Chrysler had one vehicle in 2018. Toyota followed suit in the last few years. According to the Fall Unified Agenda, in which various federal agencies publish their schedules of rulemaking activities, NHTSA expects to publish a new FMVSS 114 Notice of Proposed Rulemaking in June.

Last week, in response to the surge in amateurs stealing Hyundai and Kia vehicles with inadequate theft protection, Safety Research & Strategies submitted a petition for further rulemaking, to strengthen the anti-theft compliance test provision, which is virtually un-fail-able. The petition outlines the current Kia Boyz continuing mayhem, history of the standard, the basic and vague requirements of the current anti-theft test, and compares them to the specific and more rigorous requirements of the tests for immobilizers – which are optional for manufacturers.

We can’t say why NHTSA has abandoned FMVSS 114, but we can chronicle its descent into near-irrelevancy.

Theft Protection?

In the late 1960s, many amateur car thefts never had to break a window or jimmy a lock – they simply slipped into the driver’s seat of a vehicle with the keys dangling in the ignition and took off. Yet, joyriding posed a significant safety hazard, because thieves were not only stealing cars, they were getting into crashes and causing injuries to themselves and others — and at a disproportionately higher rate than vehicles that were not stolen. According to an April 1968 Federal Register Notice proposing FMVSS 114:

The evidence shows that cars operated by unauthorized persons are far more likely to cause unreasonable risk of accident, personal injury, and death than those which are driven by authorized individuals. Further, the incidence of theft, and hence the risk of accidents attributable thereto is increasing. According to a recent study by the Department of Justice there were an estimated 94,000 stolen cars involved in accidents in 1966, and more than 18,000 of these accidents resulted in injury to one or more people. On a proportionate basis, 18.2 percent of the stolen cars became involved in accidents, and 19.6 percent of the stolen-car accidents resulted in personal injury. The same study predicted that automobile thefts in 1967 total about 650,000; about 100,000 of these stolen cars could be expected to become involved in highway accidents. Comparing these figures with statistics for vehicles which are not stolen, the approximate rate for stolen cars would be some 200 times the normal accident rate for other vehicles. Thus, a reduction in the incidence of auto theft would make a substantial contribution to motor vehicle safety. It would not only reduce the number of injuries and deaths among those who steal cars, it would also protect the many innocent members of the public who are killed and injured by stolen cars each year

In response, NHTSA proposed to establish a theft protection rule. From the beginning, the agency’s goal was to remind drivers, via an audible warning, that they had left the key in the ignition and to make it difficult for thieves to activate the engine “within a short period of time.”

Some automakers questioned whether NHTSA was properly using its authority, or had a solid factual basis for its proposal.  Others supported the idea, but found NHTSA’s execution to be totally wrong. Nonetheless, there was consensus for the objective of thwarting the casual thief from nicking a car, and manufacturers were quite firm that they didn’t want to be forced to use a specific anti-theft technology to do so. They argued, persuasively, that technology was always evolving, and that it was necessary to stay nimble to keep one step ahead of criminals who can eventually figure out how to overcome a system.

The agency agreed that it would be best to keep the language of the technology neutral, and allow automakers to use, develop and innovate any strategy that satisfied the regulatory language. The Final Rule simply requires:

Each vehicle must have a starting system which, whenever the key is removed from the starting system prevents:

(a) The normal activation of the vehicle’s engine or motor; and

(b) Either steering, or forward self-mobility, of the vehicle, or both.

Although the final rule was not technically prescriptive, there were nine years of discussions about the best ways to accomplish this. And, even back then, the agency understood that insufficient shielding of ignition wires was a factor in the speed at which a thief could breach the ignition system. In May 1979, a constituent of then Sen. Ernest Hollings noted that cars were easier to steal due to manufacturers’ change from “a heavy metal casting steering-column (at a weight savings of only 1 ½ pounds) to a molded-plastic housing which no longer adequately protects the ignition lock.” In a 1971 docket, NHTSA wrote:

After a review of the state of the art in lock design and the difficulty of articulating performance standards for stronger locks, the NHTSA has tentatively concluded that it would be more effective to approach the problem of the susceptibility of locks to tampering by limiting the utility of removing the lock. Consequently, a new requirement is being proposed which requires the ignition systems to be inoperative if any part of the ignition lock is removed. To further protect the ignition system, the agency proposes also that the wires which activate this system shall be shielded so that they cannot be directly contacted from within the passenger compartment. The shielding could be provided by the vehicle structure or by other means. The agency is considering establishing a requirement that would necessitate the use of metal or other similar strong shielding materials which would have to be cut by special cutting tools before access to the ignition wires could be gained.

At the time, the idea that the ignition lock should be designed to resist removal was supported by major automakers such as GM, Ford, Mercedes and VW. And NHTSA noted that it was evaluating a study by the National Bureau of Standards of tensile, torque and extraction testing on current ignition lock systems to determine if it would propose specific lock retention and system operations performance standards in the future. The agency never did propose an ignition shielding requirement, and theft protection rulemaking never advanced any further.

NHTSA’s Office of Vehicle Safety Compliance (OVSC) lays out the procedures for its contracted testing labs performing FMVSS 114 compliance tests, the primary tests associated with the theft prevention aspects of rule state:

“With the key removed from the starting system, attempt to start the vehicle engine or motor. If the vehicle is equipped with an advanced key system, it may be necessary to move the electronic key device outside the minimum effective range for loading the electronic key into the vehicle starting system.”

“Determine if the steering wheel locks as a result of removing the key from the starting system by rotating the wheel in both directions.”

“Determine if forward self-mobility is prevented whenever the key is removed from the starting system. If the vehicle is equipped with an advanced key system, it may be necessary to move the electronic key device outside the minimum effective range for loading the electronic key into the vehicle starting system.”

So, test labs have to try to start a keyed vehicle without placing the metal key in the ignition slot, or, in the case of a keyless ignition vehicle, move the key fob out of vehicle range and attempt to start the vehicle by depressing the Start/Stop button, then try to  move the steering wheel and determine if vehicle can move forward under its own power.  No tools or methods for accomplishing this are prescribed. How any automaker could actually fail to comply is beyond The Safety Record’s comprehension. The procedure only partially addresses the second part of the requirement, that a FMVSS 114 compliant vehicle must also prevent “either steering, or forward self-mobility, of the vehicle, or both.”

While OVSC’s lab test procedures states they aren’t intended to limit or restrain its contractors from devising and using any techniques or equipment to obtain the required compliance test data,  and the document isn’t an endorsement or recommendation for using a particular test method, the practical reality is the opposite. Is a contracting test lab really going to devise a different test method based on an historical review of the intent of the standard? Not likely.

But maybe they should review a legal interpretation from 2004 from then-NHTSA Chief Counsel Jacqueline Glassman who recognized the intersectionality of the two-part requirement in a response to an unidentified automaker, who requested agency guidance on its engine control immobilizer module and the requirements of FMVSS 114. Glassman agreed that the system the automaker described would be compliant with FMVSS 114 because if an attempt was made to circumvent the ignition lock, including through “hot-wiring,” the immobilizer prevented engine starting without the key.

Immobilizers, defined by NHTSA as “an anti-theft device that combines microchip and transponder technology with engine and fuel immobilizer components that can prevent vehicles from starting unless a verified code is received by the transponder,” are not required, but most automakers use them.  They showed up in the 1980s and today, according to the Highway Loss Data Institute, 97 percent of MY 2021 model cars have immobilizers, compared to 26 percent of MY 2021 Hyundai/Kia models. In contrast to FMVSS 114’s meaningless compliance test, if a manufacturer chooses to use an engine immobilizer anti-theft technology, there is specific criteria that the immobilization system must be designed to meet so that it cannot easily be defeated to allow forward self-mobility by either disrupting the voltage or by using tools. These criteria enumerate 18 common tools auto thieves are known to rely on: Scissors, wire strippers, wire cutters and electrical wires, a hammer, a slide hammer, a chisel, a punch, a wrench, a screwdriver, pliers, steel rods and spikes, a hacksaw, a battery operated drill, a battery operated angle grinder; and a battery operated jigsaw

Enter the Kia Boyz. With one viral 2021 TikTok video, every issue associated with the safety hazard of quick and easy car thefts identified and debated more than 40 years ago has been borne out in the ongoing crime wave involving MY 2010-2021 Hyundai and Kia vehicles with traditional metal keys. Thieves can just rip off the steering column shroud, pull out the ignition cylinder, and using the end of a USB cable, grip the ignition cylinder to twist it, starting the engine. The whole deal can go down in about 30 seconds.

In the span of approximately a year, Hyundai/Kia’s decision to forego basic anti-theft features resulted in 14 crashes and eight deaths, according to a NHTSA press release in February. But surely, these are undercounts, because in March and April alone, the news media reported another 24 such crashes, resulting in four more deaths (including a six-month-old boy), 28 injuries, nine of which were said to be serious injuries, and one house fire, in addition to structure damage and other vehicle damage, including two police cruisers, caused by drivers of stolen Hyundai/Kia vehicles.

Hyundai/Kia now facing a demand from 17 state and Washington, D.C. attorneys general to launch a recall, and lawsuits from many municipalities, including Buffalo, Milwaukee, Rochester, Columbus, Madison, St. Louis, New York City, Cleveland, Seattle and San Diego, auto insurers, and multiple class actions. One entity not holding Hyundai/Kia to account is NHTSA. NHTSA didn’t open a defect investigation or press Hyundai/Kia to launch a formal recall with its notification and accountability requirements. NHTSA took no enforcement action at all, because it didn’t believe that the theft protection safety standard was actually enforceable. In response to inquiries from Safety Research & Strategies, a NHTSA media representative stated that the agency could not act more forcefully, because “the standard does not define normal activation.”

If that’s the case, it’s high time the agency either define normal activation or re-fashion a meaningful compliance test that addresses FMVSS 114’s historic intent.

Rollaway Prevention?

FMVSS 114 pivoted to the rollaway problem in 1988, in response to reports of crashes and injuries caused by steering wheel lock-up when the key is inadvertently removed, and rollaways caused – primarily – by children moving the gear selector from Park to Neutral. By 1990, the agency published a Final Rule requiring vehicles with automatic transmissions that have a Park position to have a key-locking system that prevents removal of the key unless the transmission is locked in Park or becomes locked in Park as the direct result of removing the key. The agency noted that this amendment accomplished several goals at once – it didn’t diminish the theft prevention aspect of the rule, it prevented steering wheel lock up if the key was inadvertently removed from the ignition, and it prevented inadvertent shifting of the gear selector, because the key’s status was now linked to the Park transmission position.

The final rule addressed the safety need for transmission shift locks, with the agency calculated that there are “roughly 400 to 800 relevant injury producing transmission lever shifting accidents each year.” NHTSA also emphasized that being able to move the gear selector out of Park without the key in the ignition posed a significant safety risk and that the agency has a “special obligation to reduce injuries involving children.”

Nothing much changed on the rulemaking front for the next 15 years. But ignition systems began to undergo a significant transformation, as automakers introduced keyless ignition vehicles into the fleet. No longer did the driver need a metal key to lock and unlock the doors, or even to start the vehicle. The key was now an invisible electric code stored in or removed from the ignition module, delivered by a plastic key fob. In a 2002 interpretation letter, NHTSA Chief Counsel Glassman warned an unnamed automaker that this new system would disrupt a driver’s relationship with the key:

We observe that if the ‘Smart Key’ device remained in the car. e.g. in the pocket of a jacket laying on the seat, a person would need only turn the ignition switch knob to start the engine. It appears to us that, with systems of this kind, there would be, in the absence of some kind of a warning, a greater likelihood of drivers inadvertently leaving a ‘Smart Key’ device in the car than with a traditional key. This is because the driver must physically touch a traditional key, unlike the “Smart Key” device, as part of turning off the engine. You and/or the vehicle manufacturer may wish to consider whether there are any practicable means of reducing the possibility of drivers inadvertently leaving their ‘Smart Key’ devices in the car.”

Glassman got it partly right – but didn’t address other serious problems that emerged with keyless ignition vehicles – which facilitated the omission of critical steps in the vehicle exit behavior sequence: driver’s exiting vehicles with the key fob, vehicle in Park, but without shutting engine off, creating a carbon monoxide hazard or shutting the engine off and exiting without placing the gear selector into Park, creating a rollaway hazard. Both hazards were largely addressed with mechanical key designs that prevented the removal of the ignition key unless the engine was off and the gear selector in Park.

The year 2005 was a FMVSS 114 watershed for a couple of reasons. First, although NHTSA still had not promulgated a rule to prevent rollaways by requiring a brake to shift interlock (BTSI), then-New York Sen. Hillary Clinton filed the Cameron Gulbransen Kids and Cars Safety Act, which would force the agency to do so. (In 2006, 19 major automakers scrambled to head off a regulation by announcing a voluntary agreement to install brake to shift interlock. That was swell of them, but in 2008, the Clinton’s bill passed, and the Final Rule was published in March 2010.)

Second, in 2005, NHTSA attempted to amend FMVSS 114 to address the advent of keyless ignition, by revising the definition of the key to include an invisible electronic code. This seemingly created an FMVSS 114 compliance problem, because these new systems allowed the driver to shut the engine off in a drive gear, and walk away with the fob. In a keyed vehicle, that metal key cannot be released from the ignition slot unless the transmission is in Park or automatically moves into Park when you remove the key. Not to worry though, NHTSA told automakers how to get around the rule in a Federal Register Notice: “We also explained that systems using an electronic code instead of conventional key would satisfy the rollaway prevention provisions if the code remained in the vehicle until the transmission gear is locked in the “‘park’ position.’”

This “cheat” immediately made FMVSS 114’s rollaway prevention provisions less effective.

Most manufacturers designed their vehicle electrical architectures so that if the driver turned off the engine without placing the transmission in “Park,” the ignition state would quietly move to the Accessory position to keep the electrical power on and the electronic “key” was maintained in the ignition module. These designs typically included a dashboard text message and/or indistinguishable chime that was allegedly intended to alert the driver to put the transmission into Park. In many cases, drivers exiting with the fob with the gear selector not in Park and the engine off, never saw the message as they turned their bodies toward the door, or failed to hear the chime or if they did, had no way to discern its meaning, and exited without knowing their vehicle was unsecured and the power was still enabled, draining the battery, to boot. Automakers resolved the dead battery complaints by installing battery saver features that would eventually turn off the power, but did little to protect the departing driver or nearby pedestrians from rollaways.

In 2009, NHTSA asked the industry to convene a SAE sub-committee to write a voluntary standard to address the human factors issues created by keyless ignitions. But the result was pathetic: it blamed drivers for manufacturers poor design choices and didn’t offer much guidance on countermeasures beyond than vague recommendations for warnings.

Dissatisfied, the agency opened a rulemaking in December 2011 to amend FMVSS 114 to address panic stops during an unintended acceleration, and the rollaway and carbon monoxide safety problems created by keyless ignitions. The proposal to solve for the latter were very loud 85-decibel warnings, if the driver exited without placing the transmission in Park or turning off the engine. Predictably, manufacturers loathed this idea, and excoriated NHTSA for even suggesting such a thing without rigorous data to determine a safety need. When NHTSA had proposed to hire a consultant to do some human-machine-interface investigations into keyless ignitions, the industry objected again. NHTSA hasn’t touched the rulemaking for a dozen years.

In 2014, apparently having forgotten that they told manufacturers how to meet technical compliance with FMVSS 114, while leaving the vehicle able to freely roll, NHTSA OVSC began testing 34 recent model-year vehicles to determine if push-button systems allowed the vehicle to be turned off in a gear other than park, or the key fob to be removed from a running vehicle with no warning to the driver, or allowed vehicles to be restarted without the key fob present. They found that you could turn off the engine with the transmission in drive, leave with the fob, and easily push the vehicle into a roll. The agency officially launched the compliance probe in January 2014 with an Information Request seeking sought a host of details related to manufacturers’ keyless ignition systems, ranging from the electronic architecture of the system, when the electronic code is purged from the system and the audio and visual telltales used to alert the driver when he or she has exited the vehicle. NHTSA also asked for complaint data and the safety information manufacturers provide to their customers about keyless systems.

After automakers explained exactly how their vehicles met the rollaway standard using the Accessory work around, the agency closed the probe with no findings of non-compliance. For example, in her report regarding potential non-compliances in Kia vehicles, the safety compliance engineer who conducted the investigation, noted: “Each vehicle was started with the push button control and the transmission selection control was placed in Drive. The starting system was deactivated with the push button control and the key fob was removed from the vehicle. We verified that the vehicle was not in Park by pushing it.”

After conversations with Kia, NHTSA’s OVSC learns: “The information and test data provided by Kia indicates the vehicles listed above meet all requirements of FMVSS No. 114. Regarding SS .2.1, if the vehicles’ starting system is deactivated when the transmission is not in Park, the starting system will be in the accessory position, the key (electronic code) has not been removed, and the transmission control is not required to be in Park. In addition, drivers are provided audible and, for some vehicles, visual warnings about the key and transmission position.”

NHTSA eventually told automakers they needn’t bother to answer the question of how many complaints they had logged.

In 2015 and 2016, NHTSA opened three separate rollaway-related defect investigations into vehicles with novel shifters, with spotty results.

The first, PE15-030, in August 2015, investigated rollaways after intended shifts to PARK in 2014-2015 MY Jeep Grand Cherokees vehicles with monostable-style shifters. The monostable is a T-handle style shifter with a similar appearance of a traditional mechanical T-handle shift selector. But monostable designs are activated by depressing a button on the handle and moving the handle rearward or forward to achieve a transmission position; the handle always returns to the centered/neutral position. The design provided no tactile feedback, so drivers trying to achieve Park were inadvertently selecting Reverse or Neutral. In April 2016, FCA brought the investigation to a swift close by recalling Jeep Grand Cherokees, Chargers and 300s with the monostable shift levers. FCA blamed drivers for failing to respond to the vehicles’ warning strategies, but nonetheless installed Auto Park, which automatically moved the electronic monostable transmission selector into Park when the driver failed to do so, and the ignition switch was pressed to turn of the engine, or if the driver attempted to exit the vehicle without the Park engaged.

The second two defect investigations were opened on the same day in December 2016, into vehicles with rotary style electronic shifters:  one covering MY 2013-2016 Ram 1500 and MY 2014-2016 Dodge Durango, and the MY 2012 through 2014 Land Rover/Range Rover Evoque and Jaguar XF vehicles. In both cases drivers complained that the vehicles had rolled after the drivers had placed the transmission in Park – or thought they did. In August 2020, NHTSA closed the investigation into the Jaguar / Land Rover vehicles, after declining complaints and claiming it found no mechanical or electronic fault that would prevent Park from being engaged; this was likely case of driver error, the agency said. While FCA initiated a series of service campaigns to add automatic shift to Park software updates, the investigation into the FCA vehicles is still open six years and five months later.  

So there you have it, after nearly 40 years of regulatory progress, FMVSS 114 begins its descent into a theft protection and rollaway prevention safety standard that does neither. Most of the time NHTSA has an opportunity to enforce it or improve it, the agency backs off, and the public is less safe because of it.

 

Amazon Delivery Drivers Sprint After Rolling Delivery Vans

Editor’s Note: This blog has been updated to reflect the software update recall initiated by Mercedes-Benz and posted by NHTSA on Jan. 10, 2022.

In August, the National Highway Traffic Safety Administration (NHTSA) opened a rollaway investigation into the Mercedes Sprinter vans with e-shifters that automatically move the transmission into Park when the driver leaves the seat. Used by Amazon’s delivery service providers, these vans have been sprinting away – with the shift indicator showing Park, and sometimes with the parking brake applied, too. Just before Christmas, Mercedes delivered a temporary fix with a Grinch-y rant against drivers. But, the preliminary evidence points to a long-established mechanical source of rollaway – a defective park pawl. Blame the driver is the manufacturers’ rollaway go-to, but will it make the investigation go away?

In 2021, Amazon Logistics – the fulfillment arm of the online retail behemoth – delivered more packages than FedEx – 4.2 billion parcel shipments in 2020, according to an analysis by Pitney Bowes. And all those billions of dish racks, books, clothing, 50-lb. bags of dog food – and every other household necessity, personal obsession, and impulse purchase – reaches America’s front doors via an army of drivers manning Amazon’s fleet, consisting mostly of Mercedes Sprinter vans.

By all accounts, the life of an Amazon delivery person is short and brutish, composed of 10-hour shifts humping as many as 40 packages an hour, while an Android device called a “rabbit” maps your route and tracks your execution to ensure that you are keeping up the pace. A December 2020, first-person, day-in-the-life-of-an-Amazon-delivery-driver story published by Business Insider, reported that a driver gets in and out of that van 200 times a day. A Reddit forum for Amazon drivers paints a fairly miserable picture of the experience – low pay, zero respect and push-to-the-breaking-point level of productivity expectations – which is apparently why the position has such high turnover.

Given the number of stops a day a driver must make and the relentless pace of delivery, Amazon needed a van that would provide good protection against rollaway, the long-time safety hazard of unintended movement in a vehicle with no driver. Rollaways can occur when the engine is on, or off, and can be caused by mechanical and software failures, or by design-induced human error.

The 2019 Mercedes Sprinter van appeared to deliver – with an electronic shifter that provides one of the most comprehensive automatic shift-to-Park algorithms The Safety Record, which has pored over scores of service literature and owner’s manuals, and examined dozens of vehicles, has ever seen. (We’ll get to that in a moment) An electronic shifter replaces the mechanical connection between the gear selector and the transmission with software that sends electronic signals from the gear selector interface to an electronic control module which relays the request to the transmission. In a 2019 Mercedes Sprinter, the shifter is a steering column mounted stalk behind the steering wheel that is pressed up or down, depending on the gear. Park is a button on the end of the lever.

 

Mercedes Sprinter vans – particularly in the 2019 model year – were rolling away with enough regularity to catch the attention of the National Highway Traffic Safety Administration’s (NHTSA) Office of Defects Investigation (ODI).

In early August, ODI opened a Preliminary Evaluation of 19,000 2019 Mercedes Sprinter vans. The Opening Resume described the potential defect as: “MY 2019 Mercedes-Benz Sprinter 2500/3500/4500 vans, configured for use as Amazon delivery vehicles or ambulances, roll away shortly after being shifted to Park using the Auto-P function.” The investigation is based on 11 consumer complaints in NHTSA’s Vehicle Owner’s Questionnaire (VOQ) database, representing eight crashes and one injury. A broader search turns up a total of 19 such reports. They have narratives like these:

The delivery driver placed the van in park on a hill and engaged the emergency parking brake.  While the engine was still running the delivery driver took a minute or two to locate his package and proceeded to step out of the driver door while the van was still running.  Just as he was stepping out of the van the driver heard several loud clicks and the van proceeded roll away and was stopped by a large tree branch that caused damage to the roof of the van.  I have had several of these vans roll away some causing significant damage to the vans, other people’s property and 2 drivers have been hurt as a direct result of this same problem.

Here’s another, with a more serious outcome:

The contact’s client was utilizing a 2019 Mercedes- Benz Sprinter 2500 for work. The contact stated while the vehicle was in park, the driver exited the vehicle however, it rolled backward, flipped over, and crushed the driver’s legs. There were no warning lights illuminated. The driver sustained road and chemical burns and medical attention was provided.  A police report was filed. There was no report of a fire or airbag deployment. The vehicle was towed to an independent lot. The local dealer was not contacted. The vehicle was not diagnosed or repaired. The manufacturer was notified of the failure and inspected the vehicle where the black box was retrieved. The contact mentioned referenced NHTSA Action Number PE21019 (Power train). The failure mileage was 965.

Or, if you prefer, you can watch a typical incident:

Rollaway is a longstanding, and diverse problem — recalls and investigations going back to the 1970s show that the root causes range from broken parts, like pawls and rods, to the effect of new technologies, such as keyless ignitions and new electronic transmission gear shift designs, on driver behavior. In the last several years, The Safety Record has been reporting on this safety problem and the solutions – you can read about them here:  

Technology Has Made Rollaways Easier: Technology Can Prevent Them

The Persistence of Rollaway

The 2019 Sprinter, a vehicle predominantly used by delivery expeditors, is the first model year of the third generation of the van, and the first year to use the e-shifter, a design found on many other Mercedes models. In 2018, Amazon announced that it had ordered 20,000 Mercedes-Benz Sprinter vans that would be leased by its Delivery Service Partners – a program for small business owners to deliver packages ordered through Amazon. Amazon Sprinter vans have an auto-shift-to-park function that automatically moves the electronic shifter to the Park position, if the driver attempts to exit without first securing Park, regardless of whether the engine is on or off. Auto-shift-to-Park features first emerged as a safety measure on passenger vehicles with e-shifters that used non-standard shift controls – like buttons, rotary knobs, and Monostable stick selectors that always return to the center position after gear selection. These new shifter designs lack the mechanical detents and consistent PRNDL sequence that most drivers were used to. This has led to drivers misjudging the gear position and exiting the vehicle without securing it in Park, followed by NHTSA investigations and recalls to implement a failsafe, such as an automatic shift-to-park feature.

Further, if the Sprinter van is not in Park and the driver opens the door, or unbuckles the seat belt or gets out of the seat, the e-shifter moves to Park. That covers all of the components of driver exit. That last condition – driver leaves seat – is determined by a driver seat occupant detection sensor, a design feature that is atypical for in these designs. While Automatic shift-to-park features in the U.S. fleet work under a variety of conditions – some only activate if the driver has turned the engine off, some don’t work if the transmission is in Neutral. Most use the combination of seatbelt unbuckled, no brake application, and the driver’s door open as the markers for driver exit. But, the Sprinter design with its driver occupant detection sensor as the trigger seems to acknowledge the realities of an Amazon delivery driver’s day.  

The Sprinter is also available with an optional Electronic Parking Brake (EPB). EPBs are an increasingly common feature that can be configured to automatically activate under whatever conditions the automaker chooses. Again, some only activate when the engine is turned off, or when the brake hold mode is manually activated, a convenience feature that holds the vehicle at a temporary stop – like a traffic light – without requiring the brake pedal to be depressed, and then releases when the accelerator is depressed to resume travel. Other manufacturers automatically apply the EPB to prevent rollaway. For example, 2014 model year Jeep and Chrysler models added “Safehold” that will engage the EPB automatically if the transmission is not in Park, the seat belt is unbuckled, the service brake is not applied and the driver’s door is opened. The commercial version of the 2019 Ram ProMaster 2500 van, a rebadged Mercedes Sprinter – also used as Amazon delivery vans – have a mechanical shift lever and an automatically applied EPB. Others with auto-applied EPBs include Ford, GM, Volvo and Mercedes – in other models.

The Amazon-configured vans have an old-school hand-operated mechanical parking brake.

So how is this happening? Drivers are reporting that the transmission was in park, and some maintained that they pulled the mechanical parking brake lever, too. Despite Mercedes’ thoughtful approach to rollaway prevention using vehicle electronics, the problem appears to lie with its mechanical components. According to David Bizzak, a Monroeville, PA-based mechanical engineer who has examined the failure mechanism in several Sprinters, the park pawl, which locks the transmission to prevent vehicle movement doesn’t consistently maintain engagement in the ring gear. This design was intended for use on lighter vehicles, but the heavier Sprinter vans appear to put greater forces on the tapered park pawl, and that added friction can force the pawl out of engagement when under load.

“This transmission has been used in Mercedes’ car line for many years, and to our knowledge there isn’t a rollaway issue with their passenger cars,” Bizzak says. “We believe it may be possible that the heavier weight of the Sprinter van creates a situation in which this disengagement of the park pawl can occur.”

In addition, says Bizzak, the parking brake requires a lot of clicks – like 10 – before it is fully engaged.

It’s difficult to know where NHTSA’s investigation is at. The only documents in the public file are the Opening Resume, and an information request to Mercedes, asking for things like its communications with Amazon, and an explanation why FedEx, which also uses Sprinters, is not having the same problem. We do know that NHTSA was present at an inspection that Bizzak also attended in April 2021 where this park pawl problem became evident.

In the meantime, on December 15, Mercedes filed a recall Part 573 Notice of Defect and Noncompliance report in which it blamed a rare condition it calls a “park lock system error” that could only result in a rollaway if the driver does not put the vehicle transmission in Park and set the manual parking brake. (NHTSA acknowledged the recall on January 7.)  The remedy involves installing revised software in the ESP control unit on the recall population.

The defect could affect as many as 48,000 of 2019 to 2021 Mercedes and Freightliner Sprinter vans, built on Platform 907, between June 05, 2018 and November 30, 2020 with a manual parking brake and a 7-speed automatic transmission that was manufactured between May 1, 2018 and October 31, 2019 in the Hedelfingen transmission plant in Germany. How many out of the 48,000 are so afflicted? Mercedes guesses it’s at 1 percent.

Mercedes mentions two problems: A “rare and temporary park lock function error” it identified during testing, and it “a separate factor” that can contribute to the park lock function error “or enhance the related rollaway risk: the ‘Park Lock Support’ (‘PLS’) function.” The PLS is part of the ESP Control Unit, not the transmission, which “applies continued brake pressure after the vehicle is stopped and the gear selector is placed in ‘P.’” Mercedes could not isolate that cause of the failure, “despite its intensive investigation, multifaceted testing, analyses, and evaluation.”

The PLS applies brake pressure after the transmission is put into Park, so it almost functions like an automatically applied electronic parking brake. The recall notice does not explain how long the PLS is active after putting the transmission into Park, or why it is even necessary for the vehicle to apply hydraulic brake pressure after the driver or the vehicle automatically shifts into Park.    

Or maybe, the PLS is more like a brake hold or hill hold – driver convenience features found on many vehicles – which, if activated, allow the driver to take their foot off the brake while at a temporary stop, like at a traffic light or on an incline. Brake Hold / Hill Hold features apply  hydraulic brake pressure to the wheels and typically release when the driver depresses the accelerator pedal and resumes travel. Brake Hold / Hill Hold systems requires a running engine to power the hydraulic brake pressure; it is meant to be temporary, and in most vehicles with the feature, the Brake Hold mode only lasts between three and 10 minutes. In many vehicles with an electronic parking brake, the Brake Hold / Hill Hold will automatically apply the EPB after a specified time period, to hold the vehicle in place, or if the driver shuts off the engine and attempts to leave the vehicle while it is being held stationary in Brake Hold Mode.

In either case, Mercedes is doesn’t disclose how it’s resetting the PLS parameters, but the essence of the system suggests that it is creating an EPB-like system using the hydraulic brakes to ensure that the vehicle will remain stationary for some unspecified time – likely the amount of time it takes an Amazon delivery person to hop out of the van, deposit a package on the doorstep and be on the way to the next stop.

The Safety Record has questions why these specific vehicles and these specific transmissions? Had Mercedes already implemented a mechanical or software fix to prevent these rollaways in the vans after Nov. 30, 2020, without filing a Part 573 within five days of deciding that there was a problem with the 2019s?    

Mercedes maintains that the driver has to follow “parking instructions” – in other words, put the transmission into park and get all 10 clicks on the parking brake lever to prevent a rollaway. Why does it matter whether the driver or the vehicle shifts the transmission into Park? According to the Mercedes service description, the Park Lock function is agnostic on the subject of how the Park position is achieved:

Park pawl control, general 

The park pawl is used as an additional safety feature for the parking brake and its purpose is to secure the vehicle to prevent it from rolling away inadvertently. 

The park pawl essentially consists of the parking lock mechanism and the electrohydraulic components located at the rear of the transmission housing. There is no mechanical connection whatsoever between the park pawl components of the automatic transmission and the DIRECT SELECT lever (S16/13) (“Park-by-Wire”).

The park pawl is engaged and disengaged purely by electrohydraulic means, either by operating the DIRECT SELECT lever or depending on various factors such as opening the driver’s door whilst the drivetrain is operational.

 

Is it really the application of the mechanical parking brake that would prevent the rollaways? Is it realistic to think that Amazon deliverers will take the time to set the parking brake scores of times a day with enough force to hold the vehicle and without running afoul of the rabbit on their phones? Why did Mercedes bother with such a comprehensive auto-park algorithm? If the parking brake is the primary park feature, why is this manual design used rather than an EPB that can ensure full clamp load and provide automatic application?

A week later, Mercedes issued a temporary fix and “an important safety reminder.” The former is an update to the PSM module which would sound a horn if the driver leaves the seat without the parking brake engaged. Mercedes says this was tested on some fleet vehicles and is effective.

The latter is one of the nastier blame- the-driver statements we have seen from a manufacturer – and we’ve seen a few. For example, in the late 1970s to early1980s, Ford JATCO transmissions that were experiencing Park-to-Reverse were linked to 306 deaths. In 1977, NHTSA initiated the first of several investigation into the root causes. It found two design flaws and evidence that Fords were at least 12 times more likely to experience Park-to-Reverse than GM or Chrysler vehicles. Ford denied that there was any technical or mechanical defect, and placed the blame on drivers:

…Ford has submitted data to NHTSA data, views, and arguments to establish that no such defect exists. Ford believes that these submissions demonstrate that unexpected vehicle movement is the result of drivers’ inadvertent misplacement of the gear selector lever, can and does occur, though rarely, on all manufacturers’ automatic transmission-equipped vehicles, and is not attributable to any defect in the vehicle design or construction. When unexpected vehicle movement incidents do occur, as they can on any automatic transmission-equipped vehicle, serious consequences may result. Ford believes, however, that such incidents can be avoided if before leaving the driver’s seat, all drivers observe three commonsense steps to make sure the vehicle is securely immobilized. These steps are: 1) properly engaging the transmission system in “park,” 2) setting the parking brake, and, 3) shutting off the engine.

More recently, Chrysler took a swipe at customers who bought one of its products with a Monostable e-shift design. The T-handle style shifter that was located on the center console between the driver and passenger, which looked like a traditional mechanical shift lever, but was activated by depressing a button on the handle and moving the handle rearward or forward – but the handle always returns back the centered/neutral position, which made it easy to misjudge the gear selection.

In April 2016, FCA recalled 811,146 MY 2014-2015 Jeep Grand Cherokees, and 2012-2014 Dodge Chargers and Chrysler 300s, to stave off a NHTSA investigation that was gathering steam.  NHTSA had tested the shifter and found that “the monostable gear selector is not intuitive and provides poor tactile and visual feedback to the driver, increasing the potential for unintended gear selection.” The remedy was the installation of an automatic shift to park feature. The recall notice scolded drivers for failing to heed FCA’s crappy warnings:

FCA US has determined that the existing strategies built into these vehicles to deter drivers from exiting the vehicle after failing to put the transmission into Park have not stopped some from doing so. Drivers erroneously concluding that their vehicle’s transmission is in the PARK position may be struck by the vehicle and injured if they attempt to get out of the vehicle while the engine is running and the parking brake is not engaged. FCA US has therefore determined that the absence of an additional mechanism to mitigate the effects of driver error in failing to shift the Monostable gear selector into PARK prior to exiting the vehicle constitutes a defect presenting a risk to motor vehicle safety.

Mercedes’ statement is more obnoxious by several orders of magnitude and makes the others look subtle. (Read the entire statement.) Here are some of the best parts:

We note that recently there appear to be drivers of Amazon’s Sprinter fleet who continue to refuse to follow important safety guidelines. Most importantly, each and every time an Amazon Mercedes-Benz Sprinter van is parked, the driver must properly shift the vehicle into “park,” set the parking brake per the instructions in the owner’s manual, and otherwise obey all local laws (e.g., turn the wheels to the curb when parking on an incline, etc.). These rules must be followed every time a Mercedes-Benz Sprinter van is parked — there can be no exceptions. Following these rules is critical to prevent vehicle roll-aways, and the risk of serious personal injury and/or property damage that accompanies them…. we continue to observe instances of drivers buckling their seatbelts improperly, in an effort to try to “trick” the system into thinking the driver is properly using his/her seat belt. For example, seatbelts are pulled behind seats then buckled or otherwise buckled outside of the proper normal fashion around the seat occupant. This not only is illegal under applicable local law, it also is terribly unsafe, and must never occur, as it can result in serious injury.

Mercedes went on to say that it has been begging Amazon to implement this horn honk.

Hmm, why wouldn’t Amazon want their branded delivery vehicles to be honking all the time, waking up napping babies, interrupting Netflix shows and generally annoying their customers? Will Amazon’s drivers follow those safety rules, including setting the manual parking brake every time with enough force to prevent rollaway, while under constant pressure to meet their insane schedules? Videos of these deliveries show that some drivers, racing to meet deadlines don’t close the van door as they dash out to leave a package.

The better question is: Why doesn’t Mercedes put a more robust parking pawl in these delivery vans loaded with packages?

 

 

Hyundai-Kia’s Billion Dollar Engine Problem that Broke the NHTSA Civil Penalty Barrier

A federal judge in California has put one class-action lawsuit in peril and approved a settlement in another alleging defects in Hyundai and Kia engines. The Korean automaker’s billion-dollar-plus legal liability is on top of the largest civil penalty the National Highway Traffic Safety Administration has ever levied against an automaker. The language of the consent agreement was opaque, but the public record is clear.

On May 11, U.S. District Judge Josephine L. Staton approved a $1.3 billion settlement that consolidated several 2017 and 2018 nationwide class action lawsuits, alleging that Hyundai Kia refused to recall vehicles with the Theta II GDI engine, even though the automaker knew the engine was defective. Representing 3.9 million owners of Hyundai Sonata, Santa Fe, and Tuscon vehicles, and Kia Optima, Rio Sorento, Soul and Sportage vehicles, roughly in the 2011 to 2019 model years, the plaintiffs charged that the Theta II GDI was prone to catastrophic failures and non-collision fires, which exposed vehicle owners to safety hazards and economic losses.

This ruling comes on the heels of Judge Staton’s dismissal last week, with leave to amend, of another California class alleging that Hyundai/Kia vehicles with Gamma 1.6L Gasoline Direct Injection engines were defective – different engine, same complaints: stalling, excessive oil consumption and fires. Judge Staton ruled that this complaint, filed in August 2020, failed to assign a cause to these symptoms other than poor manufacturing quality. 

And, six months earlier, NHTSA announced a whopper of a civil settlement with Hyundai Kia – $210 million – the agency’s largest penalty ever. It stands out because it’s been a minute since NHTSA has imposed any meaningful penalties on manufacturers uninterested in following safety regulations. The watershed years were 2014-2015, when the agency, with former Administrator Mark Rosekind at the helm, issued 10 consent orders totaling more than $530 million against a variety of manufacturers, including Hyundai, Honda, GM, and Fiat Chrysler, mostly for untimely recalls and failing to submit Early Warning Reports. But perhaps, with the Trump administration gone, it feels safe to regulate and enforce again.

(NHTSA kicked off 2021 with a $30 million penalty against Daimler Trucks North America following a 2018 investigation into seven recalls that were launched in 2017 and 2018. According to the Consent Order, the agency charged that, based on Daimler’s chronologies of events, it failed to launch timely recalls. In addition, the agency said, Daimler failed to produce timely information — including field reports — to NHTSA as it conducted a recall query. The agency opened Audit Query 18-002 in April of that year to examine four recalls that involved semi-trucks, school buses with wheelchair lifts, and recreational vehicles. Later, NHTSA expanded it to include three more recalls – two of which were expansions of earlier recalls.)

What did the Korean automaker do to earn such a righteous slap? The Consent Order only mentions “inaccuracies” and “omissions” in its communications with the agency and a failure to launch timely recalls involving more than 1.6 million Hyundai and Kia vehicles suffering from a manufacturing defect that allowed metal debris to cause premature wear of a bearing that could lead to an engine stall.

But the troubles of Hyundai Kia vehicles with Gasoline Direct Injection (GDI) engines are much more complex than that. Over the last five years, millions of Hyundai and Kia models have been the targets of four different investigations, class-action lawsuits, extended warranty programs, a product improvement campaign, and 10 recalls to deal with defects that cause stalls and non-crash fires attributed to a variety of causes. Even as Hyundai Kia agreed to pony-up to settle the class-action and NHTSA’s allegations of untimely recalls, it remains under agency scrutiny for non-crash fires plaguing many models in its fleet.

According to Hyundai, its troubles began in April 2011, when its Montgomery, Alabama, assembly plant changed the way it removed machining debris from the crankshaft of the new Theta II GDI engine. Hyundai had invested four years and $147 million developing the Theta II. Car and Driver described its launch as “the next step towards [Hyundai’s] bold goal of becoming the fuel-economy leader.” Unveiled in 2009, the 2.4-liter GDI’s first application was the 2011 Sonata.

But, once in the field, Hyundai saw its warranty numbers rise, as customers filed claims for excessive noise, an illuminated check engine light, and – to a lesser extent – moving stalls. Hyundai wasn’t overly concerned, it later told NHTSA, because the majority of those customers did not mention the speed of the vehicle at the time of the stall and they were also able to restart their vehicles and/or move the vehicles to the side of the road.

By June 2015, the complaints to NHTSA’s VOQ database were approaching heights that caught the agency’s attention, and it reached out to Hyundai. The automaker conveyed its lack of concern – a feeling NHTSA did not share, especially in the case of a high-speed stall.

In September 2015, Hyundai recalled 470,000 Model Year 2011-2012 Sonata vehicles equipped with 2.4L and 2.0L Theta II GDI engines. Hyundai described the defect as metal debris generated while manufacturing the engine crankshaft being left in the component’s oil passages. Over time, these metallic bits could be “forced into the connecting rod oiling passages restricting oil flow to the bearings,” reducing the flow of oil and possibly raising the temperatures. This condition could lead to premature wear of the connecting rod bearing, eventual failure, and a vehicle stall. The remedy for this defect was an engine noise inspection, which consisted of moving the vehicle to a quiet place and positioning a mobile tablet near the steering wheel to assess the engine sound. An algorithm – unexplained in the repair instructions – determined if the vehicle passed or failed the inspection. The latter got a new engine; the former got a new dip stick and an oil top-off.

At the same time, Hyundai extended the warranty for the engine short block assembly for all recalled vehicles, plus the net two model years that had been manufactured at the Montgomery plant – in other words, they launched a silent recall.

Kia also had Optima, Sorento and Sportage models equipped with 2.4L and 2.0L Theta II GDI engines, but it took no action in 2015 because it checked its Theta engine manufacturing process, which involved a separate assembly line using different procedures, and found no issues, with “extremely low” rates of warranty and field claims.  

Over the next 18 months, according to chronologies Kia and Hyundai submitted to NHTSA, they continued to monitor the issue. Kia contracted engine manufacturer Translead to conduct a detailed review of Kia warranty-returned engines. Translead identified an oil delivery issue, but the claims rate was still low, so Kia did nothing.

In May 2016 — Kia told NHTSA — it learned about Hyundai’s extended warranty program for the Sonatas (Really? It took eight months before they knew about that?) and took another look at the field data. The claims were still low, but picking up as customer satisfaction eroded under the high repair costs for out-of-warranty vehicle owners. Kia also decided to extend its warranty to all 2011-2014 Optima owners with 2.0 or 2.4-liter GDI engines to 10 years or 120,000 miles – in other words, it did a silent recall.

The field data for Theta GDI engine claims for the 2011-2014 Sportage and Sorento vehicles was also rising – although by not as much as the Optima. In August 2016, Kia decided to launch another silent recall covering those models. Kia encouraged owners who heard engine knocking sounds to bring their vehicles in for repairs, but dealers were turning away anyone who could not produce their oil change records.

That same month, Hyundai engineer and 26-year company veteran Kim Gwang-ho, flew to the U.S. to meet with NHTSA officials. Kim, a member of Hyundai’s Quality Strategy team, which makes recall decisions, alleged serious safety lapses involving the Theta GDI engines. Citing an internal quality report, Kim raised his concerns that the 2015 recall did not cover the entire population of affected vehicles in the U.S. and South Korea. He also stressed thatthe problem was also related to the engine design, not just a manufacturing defect.

Throughout the fall, Kia determined that its customer mailing list was out-of-date and the extended warranty notices were not reaching consumers, and it did a second mailing. By December, the VOQ data showed that people who complain about a stall still have enough motive power to get to the side of the road. Complaints were also dropping as its customers become aware of “remedy and free repair” offered in the silent recall.

Likewise, Hyundai continued to monitor engine-related field data during 2016 and into 2017. Despite having blamed the problem on a manufacturing process that it had long-ago corrected, Hyundai was tracking a rise in claims for engine replacement in later model years related to substantial noise or illuminated check engine or oil pressure warning lights, or stalls at higher speeds. By March 24, 2017, Hyundai decided to convert the silent recall for the remaining 2013 and 2014 Model Year Sonatas to an actual safety recall (17V-226), and to add the Santa Fe Sport vehicles manufactured at the Montgomery plant, which were having the same problems. Total recall population – 572,000 MY 2013-2014 vehicles.

Coincidentally, four days later, Kia launched its own recall (17V-224) for 443,825 MY 2011-2014 Kia Optima vehicles; 165,918 MY 2012-2014 Sorento vehicles; and 8,417 MY 2011-2013 Sportage vehicles, “based on anticipatory risk concerns.”

Also coincidentally, both described the defect in near identical language – metal debris from factory machining, leading to restricted oil flow to the main bearing, engine knock and an illuminated oil pressure warning light, and causing eventual failure and a stall. Neither explained why machining debris was still afflicting engine assembly lines even though Hyundai supposedly fixed it, and Kia never had the issue to begin with.

Given its inside intel from Kim, NHTSA was understandably suspicious of these unconvincing Part 573 Notices of Defect and Non-compliance, and in May 2017, opened a Recall Query to determine if Hyundai and Kia met its regulatory burden of issuing these recalls within five days of learning of a defect.

Something’s Burning

While NHTSA was digging into Hyundai Kia’s stall issues, the Center for Auto Safety was collecting instances of non-crash fires in an overlapping population of vehicles that shared the Theta II GDI engine. In June 2018, CAS petitioned NHTSA to open an investigation into the Hyundai MY 2011-2014 Sonata and Santa Fe vehicles and Kia MY 2011-2014 Sorento and Optima vehicles. The non-profit cited 120 consumer fire complaints filed to the agency’s VOQ database and another 229 VOQs of melted wires, smoke, and/or burning odors. 

The agency obliged, opening a Defect Petition investigation in August 2018 covering 2.2 million vehicles. The Opening Resume noted that the majority of the reported fires appeared to be related to the engine failures in the recall queries.

In October of that year, CAS publicly called on Hyundai and Kia to recall those vehicles, plus the 2010-2015 Kia Soul, charging that the incidents were ongoing and occurring with greater frequency and intensity. Since their initial request to NHTSA, CAS said, it had tallied an additional 103 fire reports, equaling almost one fire a day across those five models.

Litigators entered the fray, filing multiple multi-district class-action lawsuits. Under pressure from the regulators and the plaintiffs’ bar, Hyundai Kia began rolling out the recalls:

In December 2018, Hyundai/Kia launched a safety campaign involving more than 218,000 vehicles covered by the 2015 and 2017 recalls that received the engine replacement repair because the high-pressure fuel pipe may have been damaged or improperly installed, allowing fuel leaks that could lead to fires.

In February 2019, Kia recalled 378,967 2012-2016 Souls with 1.6-liter Gamma GDI engines for an overheating catalytic converter. Kia blamed faulty ECU logic for the Catalytic Overheating Protection, which could allow the high temperature of exhaust gases to damage the catalytic converter and cause abnormal engine combustion, leading to engine piston damage and rod breakage. Eventually, oil could escape the engine block, contact the hot exhaust surface, and start a fire.

It was, no doubt, a good effort. But not enough to deter NHTSA from bumping up the Defect Petition investigation to a Preliminary Evaluation (PE19-003) in March 2019. By that time, the agency counted — between its own VOQs and Hyundai/Kia reports — 1,341 non-crash fire complaints, with 26 injuries.

There is very little in the public file related to the search for root causes. But, suffice it to say, the inquiry sent Hyundai Kia in a lot of different directions. Between February and December 2020, the automaker launched 11 investigation-influenced recalls covering more than 2.5 million vehicles to address non-crash fires in a wide variety of models for a variety of causes — leaky fuel feed lines, short-circuiting ABS modules and ABS brake hydraulic electronic control units, and rod bearing failures. A December Kia recall for 295,000 Sorento, Forte, Koup, Optima Hybrid, Soul and Sportage vehicles made little attempt at a defect description. The Part 573 merely states that “an engine compartment fire can occur while driving for many reasons and depending on the severity of the fire, the identification of the cause can be untraceable.”

Yes, there are “many reasons” why NHTSA fined Hyundai Kia $210 million in November even if you can’t read about them in the consent agreement. Meanwhile, the NHTSA investigation into non-crash fires remains open.

As for the nearly 4 million Hyundai/Kia owners represented in the consolidated MDL, the $1.3 billion settlement includes free diagnostic Knock Sensor Detection Software, which continuously monitors engine performance for the symptoms that precede engine failure; a Lifetime Warranty covering all costs associated with inspections and repairs; reimbursements for any out-of-pocket costs borne by owners who had their vehicle repaired outside of or before the recalls; and loss of value payments.

The Theta II was intended to be the engine that rocketed Hyundai Kia’s cars to the top ranks of fuel efficiency. Instead, it broke federal civil penalty records. Theta, the eighth letter of the Greek alphabet, denotes a variety of mathematical and scientific concepts, such as a plane angle in geometry, an unknown variable in trigonometry, and a potential temperature in meteorology – to name a few of the easier-to-grasp representations. Within the engineering circles at Hyundai and Kia, it might as well represent the engine from hell. 

cobalt car crash

Cobalt Cover-Up

Brooke Melton died on her 29th birthday. It was the evening of March 10, 2010, and Melton, a pediatric nurse, had just finished work and was headed to meet up with her boyfriend for a celebration dinner, when the ignition module of her 2005 Cobalt slipped into the accessory position as she drove along Highway 92 in Paulding County, Ga. Melton’s Cobalt lost all power and skidded into another vehicle. She died of her injuries. The roads were wet from rain earlier that day, and the police attributed the crash to Melton driving too fast for the conditions.

The conclusions of the official report hardly closed the book for Melton’s parents. Her father, Ken, was particularly troubled. Just days before the fatal crash, she told him that her engine had inexplicably shut off while she was driving. A few weeks after the crash, GM sent Melton a recall notice for a power steering defect that could result in greater steering effort and, possibly, a crash. It would be a year before Ken and Beth Melton took their suspicions to Marietta lawyer Lance Cooper. Lawyers for the occupants of the other car were asking questions about Brooke’s 2005 Cobalt, and the insurance adjuster was scheduling an inspection – the Meltons needed their own representation. Would Cooper consider it?

The timing was inauspicious. Cooper’s firm had just returned to financial health after losing three large expensive auto liability cases that had mired his practice in debt. So, with some trepidation, and a hunch that something was amiss, Cooper, a seasoned litigator who had earned a reputation for his quiet but vigorous and successful advocacy pursing cases for victims of vehicle and tire defects, took the on what would become the most important case of his career.

Every lawyer has a few good war stories, but few have one with as much impact as Cooper’s battle against GM for installing a defective ignition switch in millions of its models and hiding that fact for a decade. In telling his own story in clear, unadorned prose, Cooper comes across as a deeply decent man whose work is guided as much by his religious faith as by his decades of experience going toe-to-toe with global behemoths.

Officially, at least 124 people died and 257 were injured in crashes caused by the low-torque ignition switch, which could slip from the run position to “accessory,” cutting off all power and disabling the airbags and all of the advanced safety systems. GM eventually recalled 30 million vehicles worldwide and paid out $2.5 billion in compensation and fines. GM executives and NHTSA officials were dragged up to Capitol Hill to explain themselves to an angry Congress.

Cobalt Cover- Up charts the twisted five-year path to justice, from the questions Brooke’s grieving parents first raised to Cooper in February 2011 to the whimpering end of the multi-district litigation’s effort in 2016 to hold GM accountable. Along the way, Cooper introduces the reader to himself and to those who provided key assists such as the late forensic mechanic Charlie Miller, who discovered a GM Technical Service Bulletin on the defect; engineering expert Mark Hood, who found that GM had changed a key switch part; and his uber-paralegal Doreen Lundrigan, who extracted valuable notice admissions out of GM’s paper blizzard.

Full Disclosure: Safety Research & Strategies consulted with Cooper on the Melton case, and SRS President and Founder Sean Kane appears in the narrative a few times. At one juncture, Kane gives Cooper critical evidence to challenge GM’s claims that it knew of no deaths or injuries associated with the ignition defect.

“I’ve known Cooper professionally and personally for decades,” Kane says. “This book –without any hyperbole – accurately portrays him as the skilled lawyer and the trustworthy and grounded person I know him to be.”

In hindsight, the switch ignition defect had been hiding in plain sight. Even as the first vehicles with defective ignition switches hit the showrooms, automotive reviewers had taken notice of the vehicles’ inexplicable propensity for sudden shut-offs. In 2005 and 2006, GM issued Technical Service Bulletins alerting its technicians to the defect of “Inadvertent Turning of the Key Cylinder, Loss of Electrical System and no DTCs.” NHTSA launched three separate Special Crash Investigations (SCI) in 2006 and 2007, all based on non-deployment of the airbag in a severe frontal crash. In each, the investigators found the ignition in the accessory position. In April 2007, the Indiana University Transportation Research Center’s Special Crash Investigation on an October 2006 Wisconsin fatal crash put the two together – noting the accessory position and linking it to the October 2006 TSB.

But NHTSA had focused solely on airbag non-deployments. And Cooper’s first working theory was that the power steering might be to blame. As Cobalt Cover-Up recounts, eliminating the power steering as a root cause, examining exemplars, following the clues, and pushing past GM’s attempts to obfuscate and mis-direct, took the right combination of perseverance and expertise. 

The story has its villains. General Motors, of course: Five months after the Meltons reached a confidential settlement, GM issued an ignition switch recall but limited it to a fraction of the affected vehicles. GM’s announcement to NHTSA mentioned no deaths, nor the fact that it had known of the problem for years. Once again, Cooper forced GM to disclose the full extent of the problem to NTHSA – and to ensure that NHTSA took it seriously, he notified the press.

Once the debacle broke into full public view, it became GM’s mission to place all the blame for its failures on the shoulders of switch engineer Raymond DeGiorgio – casting him as a rogue employee who managed to place a sub-standard switch in millions of GM vehicles without anyone’s knowledge, and years later, switch it out with a different part, again unbeknownst to any of the automaker’s numerous managers and supervisors. And, as Cooper writes, they largely succeeded. GM’s “independent” investigation by attorney Anton Valukas pinned the fiasco on DeGiorgio, garnering headlines in the popular and trade press such as Forbes “How one rogue employee can upend a whole company,” or The New York Times’ “A Fatally Flawed Switch, and a Burdened G.M. Engineer,” or Automotive News’ “How Ray DeGiorgio threw GM investigators off track for years.”

But Cobalt Cover-Up identifies many more who enabled the defect to go uncorrected for so long and derailed a full public accounting of GM’s bad acts. Among them NHTSA, which on two separate occasions, told GM officials they were concerned about a spate of crashes in which the airbags failed to deploy, yet failed to connect the dots.  

Cooper also takes some of his peers to task – GM lawyers who suborned perjury and abetted the cover-up and the lead attorneys in the Multi-District Litigation. The leadership’s failure to do the most basic vetting of their chosen bellwether cases, he writes, turned the effort to bring a full measure of justice to the victims into a flaming wreck.

Most of The Safety Record Blog readers likely know this story. It’s worth reading anyway. Established product liability attorneys will find in this book a validation of their own experiences fighting well-resourced corporations for complete discovery. For new lawyers, it’s a master class on the careful selection of experts and the painstaking attention to the details in putting together a successful automotive defects case. But Cobalt Cover-Up is a good read for anyone. For one, Cooper takes care to define the legal jargon for the lay reader. More importantly, it’s a reminder that the civil justice system can be a powerful force for good and that attributing crashes to “driver error” is easy – finding the real problem takes a dedicated and savvy legal advocate.

Cobalt Cover-Up on Amazon

 

Is NHTSA Ready to Strengthen Seat Backs?

The National Highway Traffic Safety Administration and the automotive industry have long agreed that the Federal Motor Vehicle Safety Standard for seat back strength is inadequate. And, most safety advocates argue that FMVSS 207 has nothing to do with seat performance in real world crashes.

Today, many vehicle seats are built to far exceed the demands of FMVSS 207’s 1967 static pull compliance test on an unoccupied seat, which requires that it withstand a load equal to 20 times the weight of the seat with the load applied in a forward and rearward longitudinal direction. In 2016, David C. Viano, a former General Motors scientist, now a private engineering consultant and the industry’s go-to defense expert witness, documented a steady increase in conventional seat strength and resistance to seat rotation over the past 50 years. By the 2000s, seats were significantly stronger, even though questions about their real-world performance remain.

Yet, for half a century, the battle over FMVSS 207’s effectiveness was shaped thus: Did the standard allow seat backs to collapse in rear impacts, causing serious injury to the seated occupant and any rear-seated passenger in its path, as the safety community claimed? Or did the injury data show that seat-back collapse, euphemistically dubbed “yielding” seat backs actually protect the front-seat occupant from injury, as industry maintained?

Automakers argued that “yielding seats” limited injuries in low-speed rear impacts, and that stiffer seats would result in more severe injuries – an argument based on 1950s-era seat designs with short, flexible seatbacks and ineffective head restraints. However, in real-world, higher-speed rear impacts, weak seat backs don’t yield in a controlled manner. They simply break and collapse – a result readily observed in FMVSS 301 Fuel System Integrity rear impact tests with dummies in the driver’s seat.

Industry successfully forced this false frame – much like automakers did with their claim that serious rollover injury and death was not caused by weak, collapsing roofs but by occupants “diving” into them – around the issue, effectively stymieing any solution for decades. The concept of designing a seating system that protected occupants in both low and high-speed crashes was considered a pipe dream. 

But a recent paper published by NHTSA shows that weak seats lead to serious injuries, and there are effective countermeasures. In July, NHTSA released Front Seat Modeling in Rear Impact Crashes: Development of a Detailed Finite-Element Model for Seat Back Strength Requirements, which concluded that in rear impacts, seat back dynamic rotation should be reduced to less than 35° to prevent injury to the seat occupant and occupants seated directly behind it. NHTSA commissioned the study, conducted by EDAG, Inc., to re-examine the feasibility of increasing seat back strength by using computer finite-element (FE) modelling. EDAG’s goal was to validate the FE modelling of a current vehicle front seat design that could be used with existing anthropomorphic rear-impact test dummies to study seat performance in rear impact crashes.

EDAG researchers used the 2014 Honda Accord mid-size sedan, which they had developed for other projects, to represent typical front seats. Researchers then compared their Finite Element Analysis (FEA) computer simulations to two physical tests: a more rigorous version of the FMVSS 207 test and the dynamic FMVSS 301 Fuel System Integrity high speed rear impact crash test. For the FMVSS 207 rearward seat back pull test, the researchers loaded the seat back until it collapsed, on both a manual seat and a power seat. For the FMVSS 301 test (a 55 mph moving deformable barrier with 70 percent off-set into the rear of a vehicle), researchers substituted the barrier-to-vehicle impact with a rear impact sled test, using an acceleration pulse computed from the 2014 Honda Accord model, and placed a 50th percentile male BioRID II dummy in the front and rear seats.

EDAG used the seat back strength and dynamic motion of the front seat to determine injury potential during rear impacts, finding “significant injuries on the rear seat occupant,” in the head and the knee, both of which interacted with the front seat back as it rotated rearward. For example, in the scenario in which an unbelted occupant dummy was seated behind the driver, with seat in the full down and full rear position, the front seat rotated about 40° and hit the rear seat occupant knee. “The seat back rotation observed from this study was considered high potential to cause injuries to the rear seat occupants of all types such as children and adults.”  

These results led the researchers to conclude that a 40° seat back rotation in the FMVSS 301 sled test should be considered a failure because of the injury-producing potential, ultimately recommending that dynamic rotation of the seat back be limited to 35°.

Then, EDAG researchers used computer modelling to look at injury prevention measures to the seat back, the recliner, and the seat bottom that did not add significant weight or manufacturing costs. The FEA found that modifications to the seat back parts or recliner mechanism were ineffective, because “most of the seat back dynamic rotation was caused by weakness of the seat bottom frame parts and seat mechanism. EDAG found that a gauge increase to 3.0 mm from 1.8 mm and grade change to high strength steel on the seat bottom frames yielded the performance meeting the targets.”

This research, in part, buttresses the findings of many independent researchers, such as Dr. Kenneth J. Saczalski, Carl E. Nash, and the engineers at ARRCA, who have long argued that excessive seat rotation is a serious safety hazard to the seat occupant, and those positioned behind a collapsing front seat.

“There are some excellent points in this paper,” says Saczalski. “And I’m really surprised that NHTSA said that we need to limit the amount of rearward rotation. It was nice to see after all of this time, because the [seat back standard] doesn’t address the safety of the occupant, and we’ve published many papers showing this. You’ve got to make the seat stronger, and reliable under the various levels of use.”

To understand a vehicle seat’s safety performance, it must be adjusted and its performance assessed at the low and high position, and with dummies representing the heaviest occupants –the 95th percentile male, he adds. Some seats are strong, but when challenged by crash forces with certain occupants, and with the seats in certain positions, the small linkages fail and the seat back collapses.  

Dr. Teo Forcht Dagi, a neurosurgeon and expert in brain and spine injury, notes that “the move to reconsider the seat-back standard reflects an increasing appreciation for the importance of complex dynamic factors in injury causation and prevention. Design for passenger safety must contend with the fact that the mechanisms of injury change under different circumstances, at different velocities, and under changing loads. Not only is it incumbent upon automotive designers to re-examine their assumptions as additional data become available, they must rethink their designs and their standards for adequate safety performance.”

Other support can be found in a 2009 real-world crash study by researchers from the Center for Injury Research and Prevention at the Children’s Hospital of Philadelphia, who looked at the effect of front seat back strength on the injury risk to children seated in the rear in a rear-impact crash. CHOP examined cases from 2000-2006 of 1,035 restrained child occupants under 12 years old, seated in a second-row outboard position in rear crashes, and weighted the sample to represent 10,079 children. These data came from their Partners for Child Passenger Safety Study, a collaboration with State Farm Insurance. 

The researchers analyzed the data to quantify the overall injury risk in relation to the presence of a front seat occupant and reported front seat-back deformation. Researchers found 2.3 percent of the children sustained an AIS 2+ injury; 71 percent of those crashes had a front seat occupant and 8 percent of the cases reported front seat-back deformation. For those children with reported seat-back deformation occurring directly in front of them, there was a doubling of the injury risk. The researchers, who presented the results of this survey at the 52nd AAAM Annual conference, provided the first population-based estimates of the injury risk of rear row-seated children in rear impact crash events.

At the time of its release, Kristy Arbogast, CHOP’s  Co-Scientific Director and Director of Engineering, said  “In the automotive safety community, the debate centers around mitigating injury for front seat occupants through design of the front seat, and little focus is placed on the role of seat back design on injuries to other occupants. Using two population-based samples, this study points to at least a two-fold increase in injury risk for children seated behind yielding seat backs in rear impact crashes, after adjusting for potential confounders such as crash severity.”

Similarly, Viano’s early research into seat back strength in the 1980s included tests showing how seat back rotation injured the front seat occupant. His paper, published in 1992, documented the results of his study of the influence of seat-back angle on occupant retention in a rear impact. The paper concluded that front seat occupant retention became difficult in rear impacts once the seat back angle rotated rearward more than 45 degrees from vertical and that, at 60 degrees, occupant retention was not possible – even if a seatbelt was worn.

Rear Seat Safety Diminished as Front Seat Safety Increased

This latest paper offers another strategy for increasing rear seat safety, which has long been ignored. NHTSA, as well as seat experts, automotive injury researchers and other agencies have all acknowledged this gap in occupant protection.

In January 2017, the National Transportation Safety Board, an independent non-regulatory agency which investigates crashes and makes policy recommendations, published the results of workshop on rear-seat safety it sponsored in 2016. The report noted:

 …with the focus on advancing the safety of front seat occupants through improvements in vehicle design, regulations, and crash testing, some recent studies have indicated that the protection offered to rear seat occupants is not advancing as quickly as protection for front seat occupants. Advances in front seat design and technologies have created an environment where, for some occupants, such as older children and older adults in certain crash situations, the front seat may be safer than the rear seat. This development is in contrast to the longstanding belief that the rear seat is always the safest position for these occupants.

The workshop brought together 50 scientists, researchers, engineers, and representatives of industry to discuss short-term and long-term solutions and challenges to the many factors contributing to the trend of rear-seat injury and death, including rear seat design and low adult seat-belt usage rates in the rear.

Ten years earlier, at a May 2006 government-industry meeting sponsored by SAE, NHTSA staffer Sashi Kuppa presented an analysis of frontal (non-rollover) crashes from the FARS, NASS and State Data System databases showing that the rear seats—often touted as the safest positions in a vehicle—offer inadequate protection to their occupants. Kuppa examined rear outboard positions in 1991 and later vehicles (equipped with lap/shoulder belts) and concluded that for rear seat occupants the belt was often the source of injury. Kuppa also found that for newer cars front seats were safer for restrained adults—especially adults who were 50 years old and above—as advanced restraints were added. Other crash data analyses presented by the agency similarly concluded that rear seat belts were the major cause of injuries to rear seat occupants in frontal crashes. The results of this study showed that rear-seating positions are not receiving the same attention by manufacturers as front seats, which have had increasing scrutiny and improved regulation.  

Petition after Petition, Still No Rulemaking

Will this study move the needle on long overdue rulemaking to improve rear seat safety? The history of regulatory inaction isn’t encouraging.

Researchers have been asking NHTSA to strengthen rear seat requirements for 45 years. In 1974, Nash of the Public Interest Research Group petitioned NHTSA to upgrade FMVSS 207 by adding a dynamic rear-end impact test requirement to the standard. And, in 1998, NHTSA officially acknowledged on its website that the 207 standard was inadequate:

There have been several valid criticisms of the current Federal Motor Vehicle Safety Standard No. 207 which addresses seating systems. Generally it is acknowledged that the current standard requires inadequate seat strength to insure that the seat does not fail when a car is subject to a severe rear impact.

Yet, it ignored or denied subsequent requests for a more effective regulation. In 1989, Saczalski petitioned NHTSA to amend FMVSS 207 to address the problem of inadequate seat strength and seatback failure, noting that “during rear impact the seat backs are loaded by the inertia of the occupants upper body and the current seat back requirements result in collapse of the seat back which allows the occupant to slide out from under the lap belt thus rendering the restraint system ineffective.”

Saczalski requested that NHTSA specify that the testing load be both 20 times the weight of the seat back and 20 times the weight of the occupant, and that the seat back torque criteria be increased to 56,000 inch-pounds. 

 A year later, occupant restraint expert Alan Cantor, of the consulting firm ARCCA, Inc., petitioned the agency to amend FMVSS 207 to prohibit occupant “ramping” up the seat back during seat deformation. These petitions prodded NHTSA to so some research. In 1992, it launched a Seat Back Strength project to gather information, acknowledging the current standard was inadequate to ensure that the seat does not fail when a car is subject to a severe rear impact. Despite periodic announcements that improved seat strength and head restraints was an agency priority, among its priorities, it has proposed no new rulemaking.

In August 2004, Cantor filed another rulemaking petition with attorney Larry E. Coben, this time for amendments to FMVSS No. 208 Occupant Protection requesting the inclusion of rear seat belted dummies in the dynamic crash testing requirements and the adoption of European ECE Regulation 17 which requires unrestrained cargo behind rear seats in frontal crashes.   

Coben and Cantor recommended that the agency select dummies of various sizes and adopting FMVSS No. 208 injury criteria for the head, neck, chest, and femurs. They also recommended adopting a new method of assessing abdominal injury risk. Coben and Cantor argued that applying the same injury criteria to rear seat as front seat dummies in frontal crashes was warranted, and would not cause any undue expense.

In early December 2006, the agency rejected the petition as “premature,” because it was actively engaged in a research program examining rear seat occupant protection, that included analytical and sled tests, and simulations with different-sized test dummies in the rear seats to determine advanced restraint system feasibility and improved restraint geometry.  

In 2015, after nearly a decade of NHTSA apathy, Cantor and four of his colleagues at ARCCA Inc., re-petitioned NHTSA to amend FMVSS 207:

Since 1989, hardly a day passes that we are not faced with an issue related to seat failure in a rear-end crash and the resultant serious injury that such a failure causes. As automotive seating and restraint experts here at ARCCA, we have been involved in hundreds of seat back failure litigation cases, the vast majority of which have settled prior to trial. As part of our work on these cases, we have had the opportunity to review seat strength data from various auto makers, and we have been involved in the conduct of a  variety of both static and dynamic tests on the failing seats as well as on seats that can withstand the types of forces typically seen in today’s passenger vehicles that are involved in rear-end crashes. In addition, we have seen both the published and non-published research and data from many others, including most vehicle manufacturers. What is clear from all of this is that automotive seats are more than just “chairs” to allow people to comfortably drive cars or for passengers to be transported in luxury: seats are also safety devices that provide restraint and, in a rear-end crash, the seatback should afford the same kind of protection to the user that a seat belt provides in a frontal impact. 

The petition pointed out that the current test – using a static load in an empty seat – ignored the laws of physics: occupants of different weights or mass will result in different loading of the seatback under the same rear-end crash conditions. ARCCA recommended that NHTSA establish a FMVSS 207 dynamic test – using the FMVSS 301 crash test as a guide – and a New Car Assessment Program (NCAP – NHTSA’s five-star rating system that manufacturers use to tout the safety of their vehicles and that consumers use in making buying decisions) rear-impact test. Or, more simply, restore paragraph S4.1 (b) of FMVSS 209, which required the lap belt portion of the seat belt to remain on the pelvis under all crash conditions.  

In 2016, The Center for Auto Safety petitioned the agency to modify its “child seating recommendations by adding the following or similar warning language and that such language be required in Owner’s Manuals under 49 CFR § 571.208 S4.5.1(f): If Possible, Children Should Be Placed In Rear Seating Positions Behind Unoccupied Front Seats. In Rear-End Crashes, the Backs of Occupied Front Seats Are Prone To Collapse Under the Weight of Their Occupants. If This Occurs, the Seat Backs and Their Occupants Can Strike Children in Rear Seats and Cause Severe or Fatal Injuries.”

These petitions are likely to meet the fate of their predecessors, dying quietly in obscurity.

Improving Rear Seat Safety through NCAP

The one action NHTSA has been willing to consider is to add a rear impact test to the NCAP, and let the universe of scarce stars scare the manufacturers into either improving their seat designs, or risking their markets. April 2013, the agency published a request for comments on possible changes to the New Car Assessment Program, and noted the need to improve rear seat safety.

In recent years, improvements that have been made to the front seat crash environment have significantly decreased the risk of injuries and fatalities for front seat occupants involved in frontal crashes. While exposure and injury rates for rear seat occupants overall are still relatively low, there is an emerging need to further understand the rear seat environment in recent model year vehicles, particularly in consideration of lighter and more compact vehicle designs. Comments are requested on the availability of any data that illustrate whether safety benefits can be realized through encouraging additional safety improvements and/or technologies including rear seat belt reminders targeted at protecting the rear seat environment.

The agency held out the possibility of dynamically testing rear seats and seat belts in the NCAP frontal crash tests, with a 5th percentile adult female Hybrid III dummy in the rear seat behind the ATD in the driver’s seat. The notice attracted nearly 60 commenters, many of whom generally supported the idea of enhancing safety for rear seat occupants and for additional NCAP tests with the 5th percentile ADT in the rear, but disagreed over the test parameters.

The agency, however, did not move forward with rear seat dynamic tests. Instead, it announced in January 2015, it planned to include NCAP ratings pertaining to automatic emergency braking systems.

In May 2016, the trade press, from Automotive News to the Insurance Institute for Highway Safety’s Status Report, reported that NHTSA planned to add a new oblique frontal crash test, simulating two mid-sized vehicles crashing at 56 mph, with a 50-percent overlap to measure how well vehicles protect people in an angled frontal crash. The test, using a stationary vehicle and a moving barrier, would feature a THOR 50th percentile male dummy in the driver seat and the modified Hybrid III 5th percentile female dummy in the right rear. The agency planned to implement this new test in time for the 2019 model year. And the impetus for this change was not to protect children, who have heretofore been frequent rear seat passengers, and vulnerable to injuries and deaths, but to protect adult passengers using ride-hailing services, such as Uber and Lyft.

This apparently did not happen, because in 2018, the agency published a notice of yet another  public meeting to gather input from stakeholders to discuss New Car Assessment Program. The October 2018 meeting, in part, “focused on crashworthiness strategies” for NCAP, including “rear seat occupant protection.” Last month, The Safety Record Blog, with no real confidence that the agency would be able to manage a “Yes” or “No” answer, asked if NHTSA ever made good on its 2016 promise. After nine days of e-mail exchanges, a NHTSA spokesman sent us a link to a press release announcing that NHTSA would publish another notice sometime in 2020, unveiling its proposed changes to the NCAP. Would these changes include a rear seat dummy in a frontal oblique crash test? Final word from the public information specialist: “The agency will determine the upgrades for the NCAP test after its careful review of the comments and feedback submitted to the 2020 Federal Register notice.” Well, that clears everything up.  

Despite this sad, long history of agency inaction, Saczalski is happy to see the agency publish a paper acknowledging the problem of weak seat backs and using Finite Element Analysis, a valuable engineering design tool, to build a better seat.

“This the paper is very good with FEA, you can find the most appropriate design, go to prototype part and run the tests. Then you know how to tweak the system. But this process has been so slow and the children who have died over all these years because of it – it’s a significant number.”

Will the Regs Catch Up to Vehicle Autonomy?

In July, the California Utilities Commission granted Waymo (formerly the Google self-driving car project) the state’s first permit to test its driverless vehicles without safety drivers on public roadways. And, by the end of this year, the company planned to launch a driverless taxi service in Phoenix. Ford has promised the public a “fully autonomous vehicle in commercial operation” by 2021. Tesla, which has led the bumpy path on semi-autonomous vehicles, has forecasted the introduction of as many as a million Tesla “robo taxis” on the road by the end of this year.

The generally accepted wisdom that driverless cars are the future and the future is now, has presaged an influx of investor dollars and ambitious plans to level up the fleet of vehicles with no driver controls. The optimism has been as unfettered as the regulatory landscape – which is to say that this Wild West has no sheriff.

In 2016, the National Highway Traffic Safety Administration announced that its approach to this technological transition would be the light hand of guidance, and issued a list of vague conceptual statements. Last October, the Department of Transportation released its third iteration, Preparing for the Future of Transportation: Automated Vehicles 3.0. This brightly-colored, substance-free piece of public relations affirms NHTSA’s Orwellian view that although it has the authority to regulate automated driving systems, it shouldn’t. Check out some examples of NHTSA Newspeak in AV 3.0:

“The right approach to achieving safety improvements begins with a focus on removing unnecessary barriers and issuing voluntary guidance, rather than regulations that could stifle innovation.” (Fewer regulations equal more safety.)

“ADS developers are encouraged to use these safety elements to publish safety self-assessments to describe to the public how they are identifying and addressing potential safety issues.” (Industry-promulgated transparency is truth.)

“Delaying or unduly hampering automated vehicle testing until all specific risks have been identified and eliminated means delaying the realization of global reductions in risk.” (You must put yourself at risk to reduce risk.)

Indeed, the document is remarkable for its lack of attention to the most basic mandatory protections for the motoring public.

This summer, the agency sought comments for a rulemaking to exempt driverless vehicles that lack traditional human-machine interfaces, such as steering wheels or brakes from the crash avoidance (100 series) of Federal Motor Vehicle Safety Standards which regulate those components. Specifically, NHTSA was seeking public comment on the near- and long-term challenges of testing and verifying the compliance of automated driving systems (ADS).

The Advance Notice of Proposed Rulemaking grew out of requests by Google and GM to reconcile safety standards written for traditional motor vehicles with driverless vehicles (DV). On February 4, 2016, NHTSA responded to several Google’s concerns about how it could certify a vehicle that does not include manual controls, such as a steering wheel, accelerator pedal, or brake pedal. The response also provided tables listing those standards that NHTSA could interpret Google’s ADS as the “driver” or “operator,” and a table listing those standards that NHTSA could interpret the human occupant seated in the left front designated seating position as the ‘‘driver.’’ The agency interpreted the term “driver” as applying to the ADS.

In January 2018, GM filed a petition seeking an exemption so it could run 2,500 Zero Emissions Automated Vehicles on some undisclosed roads, and still meet FMVSSs, despite having no driver or driver controls. GM categorized the FMVSSs as those designed to interface with a human driver, such as manual controls; those that provide human drivers with information, such a telltales and indicator lamps; and features to protect human occupants, such as air bags, and argued that “its ADS–DVs without traditional manual controls require only the third category of requirements.”

Based on the issues Google and GM raised, the agency noted in the ANPRM that it was considering four different regulatory approaches: keep an FMVSS if the control was necessary for the safety of all vehicles, even if it means redundancies on DVs; axe it if the requirement is no longer necessary for any vehicle; keep some FMVSSs required for traditional vehicles only; and write separate  different control or equipment requirements for ADS–DVs. NHTSA also asked a series of questions related to the pros and cons of these approaches and about compliance testing.

The docket attracted nearly 100 commenters, including the usual suspects – industry groups, such as the Alliance of Automobile Manufacturers and Global Automakers, individual manufacturers, such as Ford and GM; and safety advocates, such as the Center for Auto Safety. At the same time, a bi-partisan, bi-cameral Congressional group has been seeking input from various stakeholders in advance of writing autonomous vehicle legislation.

Safety Research & Strategies, with its long history of safety advocacy, has submitted comments to both groups covering three topics: the lack of functional safety standards for critical vehicle controls; the lack of updated standards on the human-machine interface (HMI) of vehicle controls; and the lack of accessible data and interpretation tools to adequately monitor and identify vehicle systems for potential malfunctions. 

You can read them here: SRS Comments to NHTSA Docket 2019-0036 and SRS Comments to the U.S. House Committee on Energy and Commerce and the U.S. Senate Commerce Committee        

We argue that the problems that more vehicle autonomy will bring can already be seen in the current vehicle electronic failures and automakers’ poor human-machine interface designs. For example, the advent of keyless ignition vehicles with push button Start/Stop is resulted in unintended consequences: carbon monoxide poisoning, rollaway crashes and easy thefts – hazard scenarios that were previously eliminated under the FMVSS 114 Theft Protection and Rollaway Prevention requirements applicable to traditional metal keys.

The lack of a functional safety standard for electronic controls results in scenarios in which a critical system intended to save lives can actually create a new hazard that can take lives. For example, in May, Fiat-Chrysler recalled 4.8 million 2014 to 2018 Chrysler, Dodge and Jeep models because of an electrical short circuit that prevents the driver from manually shutting off the cruise control or disengaging it with the brakes, resulting in the vehicle maintaining its current speed or even accelerating.

The current opacity of vehicles’ internal diagnostic and operational data is another huge problem, because it hinders outside entities’, such as NHTSA or consumers, ability to independently examine, document and identify potential vehicle-related failures.  

As a vehicle takes over most of the operational functions, the amount of data it must gather, assess, and store, and the speed at which it must process this information will increase exponentially. Currently, autonomous test vehicles “typically generate between 5TB and 20TB of data per day, per vehicle.” Even in current Level 2 vehicles (defined by NHTSA as “partial automation” the vehicle has combined automated functions like acceleration and steering but the driver must remain engaged in the driving task and monitor the environment at all times.”  – think Tesla’s Autosteer feature.) the amount of data that is transmitted between modules, which is stored to widely varying degrees amongst vehicles, is extraordinary, and the tools available to the public, law enforcement and diagnosticians are generally limited to OBD II diagnostic scans and Event Data Recorders.

This has already led to motorists’ being charged civilly and criminally for at-fault crashes without the ability to properly defend themselves. Despite the plethora of data circulating in a vehicle, it may not be recorded unless a preset active fault is flagged. Further, the publicly available tools used to examine the vehicle and driver behavior, which include scan tools to extract the data from the Event Data Recorder, are able to access only a fraction of what may be needed or available to the manufacturer.

For sure, NHTSA’s hands-free approach to steering the revolution has its fans. Global Automakers argued that a safety standard, such as FMVSS 103 (Windshield Defrosting and Defogging Systems), could be dropped because it “is intended to address forward visibility for the human driver, as measured from a specified eye point at the “driver’s” designated seating position. In this case, the availability of defroster/defogger systems becomes more of a customer satisfaction issue in these vehicles, which should be left to manufacturer discretion to address.” Yeah, everyone wants to ride in a vehicle where you can’t see out of the front windows. And if the ADS’s steering system fails and the vehicle is headed for a tree, who wants to see that?

But many more commenters to the NHTSA docket pointed out that the agency’s desire to forge ahead left wide gaps in implementation. There has been little to no discussions about so many aspects of this complex evolution – among them, the interconnectedness of autonomous vehicles and the roadways they traverse, raised by the American Association of State Highway and Transportation Officials (AASHTO), and state drivers’ licensure laws, raised by the American Association of Motor Vehicle Administrators (AAMVA).

They also offered pointed criticisms of the current safety assurance process of test drives. Former Lockheed Martin systems engineer Michael DeKort, who, in 2008 won IEEE’s Carl Barus Award for Outstanding Service in the Public Interest for his efforts to expose safety and security problems with the U.S. Coast Guard Deepwater Project, chimed in to “strongly urge NHTSA to look past the hype and to the aerospace, DoD and the FAA regarding proper development and testing due diligence of not only the autonomous vehicle system but the use and qualification of proper simulation. This as a tenable and safe alternative to the untenable and reckless process of public shadow and safety driving being used now by most driverless vehicle makers.”

It is impossible to drive the one trillion miles or spend over $300B to stumble and restumble on all the scenarios necessary to complete the effort. In addition, the process harms people for no reason. This occurs two ways. The first is through handover or fall back. A process that cannot be made safe for most complex scenarios, by any monitoring and notification system, because they cannot provide the time to regain proper situational awareness and do the right thing the right way, especially in time critical scenarios. The other dangerous area is training the systems to handle accident scenarios. In order do that AV makers will have to run thousands of accident scenarios thousands of times. that will cause thousands of injuries and deaths. The solution is to replace 99.9% of that public  shadow and safety driving with aerospace/DoD/FAA simulation technology and systems/safety engineering practices. (Not the gaming architecture-based systems most are using now. That technology has critical real-time, model fidelity and loading/scaling issues. These will cause improperly trained systems, false confidence and real-world   tragedies.)

Strange bedfellows Advocates for Highway and Auto Safety and the National Automobile Dealers Association challenged NHTSA’s premise that regulations are barriers, and that driverless cars shouldn’t be subject to certain safety standards because they reference actions by human drivers. From NADA:

Proposed changes to various FMVSS to accommodate ADS-DVs should preserve the safety purpose of those FMVSS. For example, while an ADS-DV with fully automated steering may not need a steering wheel to safely navigate the roads, the ADS-DV should be able to maintain at least the same level of steering performance as an experienced and  well-trained human driver operating a vehicle with a steering wheel.

Science-for-hire behemoth Exponent made a similar point: “To ensure demonstrable equivalence between any new certification approach to current performance requirements, there must be a scientific or engineering linkage to assure vehicle level performance characteristics equivalent to existing FMVSS requirements for conventional vehicles driven by a human.”

Beyond the docket, embedded systems experts have raised concerns about the rapid adoption of autonomous technology outside of any required safety protocol. For example, where is the discussion of fail-safes? When autonomous technology goes awry, will the human occupants have an intervention mechanism? Dr. Philip Koopman, co-founder and CTO of Edge Case Research, an autonomous systems safety consulting company, and a professor at Carnegie Mellon University, regularly discusses the issues and conflicts presented by vehicle autonomy, including the unrealistic expectations of human drivers in driverless cars. From his blog Safe Autonomy:

High-end driver assistance systems might be asking the impossible of human drivers.  Simply warning the driver that (s)he is responsible for vehicle safety doesn't change the well-known fact that humans struggle to supervise high-end autonomy effectively, and   that humans are prone to abusing highly automated systems.

One doesn’t have to move far beyond the happy promises of industry and its primary cheerleader, NHTSA, to see that their confidence in a carefree transition to driverless vehicles is not based on really anything. Maybe that’s why public confidence in automotive technology is low. A 2019 Ipsos/Reuters poll found that “half of U.S. adults think automated vehicles are more dangerous than traditional vehicles operated by people, while nearly two-thirds said they would not buy a fully autonomous vehicle.”

 

Toyota Has the Most Keyless Ignition Related Deaths, But Takes no Action

EDITOR’S NOTE:  Five days after publication of this article Toyota announced its plans to add auto shutoff and auto park features for 2020 models to prevent CO deaths and rollaways – 13 years after first death and 7 years after Ford and GM.

Last month, Dr. Sherry Hood Penney, 81, and Dr. James Livingston, 88, died of carbon monoxide poisoning. The couple had inadvertently left their keyless ignition 2017 Toyota Avalon running in the attached garage of their Sarasota condo. The car ran until it was out of gas and its battery died.

Penney and Livingston were distinguished in life – Dr. Penney’s resume included glass-ceiling shattering stints as interim president at SUNY Plattsburgh, the first woman at SUNY to serve as vice chancellor of academic programs, policy and planning at SUNY, and interim president of the UMass system. Dr. Livingston was a retired MIT professor, a research physicist at General Electric, and a global expert on magnets.

They died of indifference.

Toyota has the most keyless ignition carbon monoxide deaths. It had the first publicly acknowledged deaths and, now the most recent deaths. Yet, Toyota has done nothing to implement a simple, inexpensive software solution that some other major automakers introduced seven years ago.

No Manufacturer Knows the CO Keyless Ignition Hazard Better Than Toyota

The Penney-Livingston deaths bring the total of the known Toyota keyless ignition carbon monoxide fatalities to 17, with an additional 18 CO injuries. Deaths linked to keyless Toyota or Lexus models now account for 47 percent of the 37 known deaths.

Carbon monoxide deaths associated with keyless ignition vehicles represent the tip of the safety problem. The vast majority of keyless ignition vehicle owners have accidentally left their engines running at one time or another, but most incidents result in a metaphorical knock to the head – “I can’t believe I didn’t turn off my engine!” moment. In order for his error to cause injury and the death, the vehicle must be parked in an enclosed space, adjacent to an occupied living space, it must have a full, or nearly full, tank of gas, and there must be a stretch of time when interruptions from others outside the home are unlikely.

Since these deaths and injuries are not tracked in any methodical way, and because not all keyless ignition carbon monoxide deaths make the news, it is likely that there are more deaths and injuries than are publicly known.

The first known Toyota keyless CO deaths occurred in April 2006, David Colter, 89, and his 70-year-old wife Jeanette, of Port St. Lucie, Florida, were found dead of carbon monoxide poisoning in their home, after leaving their 2006 keyless ignition Avalon running. According to their son-in-law, Jeanette Colter had apparently already had one incident in which she left her Toyota running in a parking lot; and family members warned her about forgetting to turn off the car with the quiet-running engine.

“That car was so quiet, it was hard to tell if it was running or not,” Terry Wilson said in a news article. “You don’t even need a key to start it. You just push a button to turn it off and on.”

Three years later, Mary Rivera, a former college professor, left her Lexus ES 350 idling in a garage beneath her homes. Her 79-year-old partner, Ernest Codelia Jr., died of carbon monoxide poisoning. Rivera survived, but suffered permanent brain damage. The incident hit the news in November 2010, when the victims filed a civil lawsuit against Toyota. By then, there was another Toyota keyless ignition death and injury. In August of that year, 29 year-old Chasity Glisson of Boca Raton, Florida, parked her Lexus IS250 in garage to make room for boyfriend’s vehicle, and forgot to turn off the engine.  She later collapsed in bathroom on the 3rd floor, where her boyfriend, Timothy Maddock, found her, and passed out himself. The pair was found the next day; Glisson died of carbon monoxide poisoning. Maddock was critically injured and hospitalized for ten days.  An investigation revealed that the carbon monoxide came from the Lexus in the garage, which Glisson inadvertently left running.

In 2010, Safety Research & Strategies reached out to the National Highway Traffic Safety Administration’s chief counsel’s office and rulemaking engineers to alert the agency to the  rollaway and carbon monoxide hazards of keyless ignitions.

In 2011, NHTSA raised the profile of Toyota incidents in a Notice of Proposed Rulemaking to amend Federal Motor Vehicle Safety Standard 114, which regulates key systems. In a half-baked attempt to address the carbon monoxide hazard, the agency proposed mandating louder warnings. In explaining its rationale, the NPRM specifically made examples of the Glisson incident and a consumer complaint involving a close call from a 2007 Lexus LS460 owner, who inadvertently left his vehicle running in his garage, but was awakened at 2 a.m. by a carbon monoxide detector, and averted serious injury. The rulemaking was roundly criticized by industry and advocates alike, but for different reasons. Industry was opposed to the louder warnings solution for fear of annoying customers, chided the agency for proposing a rule based on consumer complaints, but also tried to block it from conducting some human factors research that would serve as the basis for rulemaking. Consumer advocates argued that auditory warnings were much less effective than a simple automatic engine idle shutdown mechanism and the agency had no data to support the effectiveness of the proposed audible alert.  (NHTSA has not advanced the rulemaking since.)

Since then, Toyota owners have suffered injuries and fatalities from carbon monoxide poisoning at regular intervals. In the first five months of 2019, there have been three known deaths caused by Toyota keyless ignition vehicles.

In February, Russell Fish, a 68-year retired member of the U.S. Army, and the National Security Administration from Hanover, Pennsylvania, and the family dog, Angel, died when Fish parked his wife’s 2011 keyless ignition Toyota 4Runner in their garage and inadvertently left the engine running. As Fish retired to his bedroom, the vehicle continued to run for nine hours until the temperature in the garage rose to levels high enough to explode the aerosol paint and insecticide cans on the garage shelves; the dash warped and the grill of the vehicle melted; and Fish slipped into death from carbon monoxide poisoning. He was found the next day by the family’s pastor, who they had check on Fish when he didn’t answer his phone.

“My research has uncovered this has been a known issue for some length of time and Toyota has had (and failed to take) the opportunity to do the right thing by correcting it,” says Fish’s oldest son, Nathaniel Fish of Phoenix, AZ. “They have failed to ensure that no other parent has to sit their children down as they’re getting ready for a party and tell them that the last time they saw their pappy was the last time; no pastor will need to find the body of one of his parishioners lying in bed in a stiflingly hot house filled with exhaust; no wife will need to get that call, while visiting grandkids, that her husband and life partner won’t be there to embrace her and welcome her home when she returns; no grandson will have his senior trip forever marred by the memory of receiving the startling news that his pappy had died from something that could have been prevented.”  

Older Drivers, Ingrained Habits

Drivers of every age can and do inadvertently leave keyless ignition vehicles running.

The design makes it easy to do. In a keyless ignition system, the key is no longer a physical object. It is an invisible code, transmitted via radio waves to and from the plastic fob and the ignition module in the vehicle. The fob must be in the vehicle to start the engine – much like a traditional metal key, but unlike a metal key, it plays no role in turning off the engine. Driver’s must execute series of actions – shift the vehicle’s transmission into PARK, push the START/STOP button and open the driver’s door, to fully remove the “key” from the ignition. Drivers exiting their vehicle with the keyless ignition fob – which manufacturers repeatedly call the “key” – believe that because they have the “key” the engine is not running. And because the fob is a proximity device that can only start the vehicle when it’s inside of the occupant compartment, many drivers believe the reverse is true – the car won’t continue to run – and if it did it would time out, like other features in vehicles including the radio. 

With a traditional metal key, a driv er cannot remove the key from the ignition without shifting the transmission into PARK and turning off the engine. Thus, the presence of the physical key outside of the vehicle provides drivers with the status of the vehicle – the engine is off, the transmission is locked in PARK – it’s a strong visual and tactile reminder. If your key was in your pocket or hanging from a hook in your kitchen, there is no way the engine could still be running or the transmission in a gear other than PARK. Those possibilities had been engineered out by regulation. That is not the case with a key fob. You could be miles away with the fob in your hand, and your vehicle could be idling away or in still in DRIVE.

It’s obvious that manufacturers conducted no human factors testing to see how drivers would perceive the new system, and what human errors the design itself would induce, because automakers did next to nothing to transition drivers to this radical shift in their relationship with the new-fangled “key.” Rather, manufacturers exacerbated the propensity for error by telling vehicle owners that the plastic fob was the key, branding their systems with names like “Smart Key” or “Intelligent Key” and depicting the fob in owners’ manuals and on the instrument panel as the “key.” They relied on auditory and visual warnings that were often never heard or seen by the driver, or, in the former case, using tones that were like other in-vehicle telltales that failed to distinguish the hazard and status of the key and lasted only seconds.

Combine these factors with today’s nearly silent engines and dashboards, headlights, radios and other electrical accessories that stay enabled for a short period after turning off the vehicle, and it is now super-easy to leave a vehicle running.

Now let’s suppose that you used a traditional metal key for the vast majority of your decades as a driver. Your key behaviors are pretty ingrained. Throw in some hearing loss and background noise, like a garage door going down to further mask that quiet engine and what are the odds you might make an error?  

Toyota Sure Loves the Mature Market’s Money

The Livingston/Penney vehicle was a 2017 Toyota Avalon, intentionally purchased, according to Mrs. Livingston’s son Michael, because “it was such a safe car.” Safety, along with quality has long been key to Toyota’s brand, and the Avalon, its full-sized sedan has long been a favorite of older car buyers.  

Eight of the Toyota CO deaths occurred in five incidents involving an Avalon.

In 2005, the year before the Colters died, “the median age of the Avalon buyer was just under 60, an age that used to signal that a consumer group – and the brand they were buying – had one foot in the grave,” according to a story published in Ward’s Auto.

Rather than disdain the aged buyer, Toyota embraced them. That same story quoted Don Esmond, Toyota Motor Sales senior vice president and general manager:

“That’s all changed now, says Esmond. “While the goal of most auto manufacturers these days is to lower the median age of their buyers as much as possible, we are quite content with the mature age profile of the Avalon buyer. To be honest, I don’t think I would have been comfortable with that number, when we launched Avalon, 10 years ago.”

Over the last 14 years, the average age of an Avalon buyer went up. Today’s “average Avalon buyer today is 66 years old, a few years older than the segment average,” according to a Forbes story published in April 2018. But, again, Toyota is bullish on older buyers. While many industry observers have written the obituary of the full-sized sedan, noting that younger buyers favor SUV crossovers, the April 2018 Forbes story marvels at Toyota’s counter-punch in unveiling a new 2019 Avalon: “As sales in the segment are sinking, several other automakers are discontinuing their big sedans and the average large car buyer is 64 and not getting any younger. Thanks to the huge baby boom populace, it will be some time yet before the market for big sedans disappears. Toyota is betting that with the redesigned it can keep its core Avalon audience happy and attract buyers who previously favored other models.”

And Toyota is supposed to be making its vehicles more age-friendly via the partnership between Toyota’s Collaborative Safety Research Center and the MIT AgeLab, which studies things like “understanding how drivers respond to the increasing complexity of the modern operating environment.”

Apparently, that effort doesn’t extend to preventing carbon monoxide deaths its keyless ignition system vehicles.

Nathaniel Fish says part of his disappointment in Toyota is the irony of its marketing keyless ignition systems to older consumers.  

“In fact, in show rooms they actively market this push button technology as a feature for older drivers who might struggle with inserting a key and turning it…to a demographic that is already conditioned through literally decades of driving cars that require a key to both turn their vehicle on and turn their vehicle off, a demographic that is often suffering hearing loss and cannot hear the increasingly quiet engines still running as they exit their vehicle while the noisy garage door is closing, to think of their “key fob” as just an easier key, without which they cannot start their car,” he says. “And so, reasonably, they view this fob as a key and treat it as such when they take it with them as they leave their vehicle, believing the vehicle cannot run without the ‘key fob’ nearby

It Will Take an Act of Congress

In February, U.S. Senator Richard Blumenthal (D-CT) introduced the Protecting Americans from the Risks of Keyless Ignition Technology (PARK IT) Act, which would require NHTSA to amend and finalize the 2011 proposed rulemaking a rule that vehicles automatically shut off after a period of time to prevent carbon monoxide poisoning, and a rule that sets a performance standard to prevent rollaway.

Blumenthal’s bill (S.543) now idles in the Commerce, Science and Transportation Committee, awaiting a Republican sponsor before Chairman Roger Wicker (R-Miss.) will advance it. Meanwhile, U.S. Representatives Jan Schakowsky (D-Ill.), Darren Soto (R-Fla.), Seth Moulton (D-Mass.) and Joseph Kennedy (D-Mass.) introduced a House version (HR.3145) of the PARK IT Act yesterday.

Specifically, to prevent rollaway the PARK IT Act would require the Secretary of Transportation, within two years of enactment,  to issue a final rule requiring manufacturers to install technology in motor vehicles equipped with keyless ignition devices and automatic transmissions to prevent movement of the motor vehicle if  the transmission of the motor vehicle is not in the park setting; the motor vehicle speed is low enough to establish that it is stationary; the driver’s door is open and driver’s seat belt is unbuckled; and the service brake of the motor vehicle is not engaged. This is the same algorithm Fiat-Chrysler Automotive employed when it recalled Jeeps and other Chrysler vehicles to add updated software to prevent rollaway.

“Basic safety standards and technology can protect owners of keyless ignition cars from the threat posed by carbon monoxide poisoning and rollaways. NHTSA’s failure to act is indefensible and has tragic and fatal consequences. Congress must move swiftly to pass the PARK IT Act and compel NHTSA to do its job,” said Senator Blumenthal.

Accompanying Blumenthal at his April announcement in Hartford, was Suzanne Zitser.  In January 2013, Zitser, a Connecticut attorney, wrote a heartfelt letter about the carbon monoxide death of her 86-year-old father, Gerald Zitser, who owned a 2006 Avalon, to any entity she thought might be interested – the U.S. Department of Justice, the U.S. Consumer Product Safety, and Toyota.

Gerald Zitser, she wrote, still rode his bicycle around town. He loved his Yankees and his Toyota. He took meticulous care of the car, and never missed an opportunity to follow his favorite baseball team. On June 28, 2012, Gerald Zitser did his shopping at the local Publix, parked his Avalon in his garage, and took his groceries into the house. He closed the garage door, but inadvertently failed to shut off his engine. Later that evening, Zitser settled into his recliner to listen to the Yankees take on the Chicago White Sox. He never woke up. (The key fob was found in his pocket.)

“Had there just been an automatic shut off system that kicked in after a preset time when their [sic] was no weight in the driver’s seat, much like the airbags in the passenger side, then this senseless, tragedy would have never occurred,” she wrote.

His youngest daughter received replies from the DOJ, the CPSC and Toyota. The latter response included a form letter assuring her “It is through correspondence such as yours that we are able to continue to improve our services.” She got a case number, and a second round of correspondence from Toyota in which the company expressed its willingness to conduct a vehicle inspection, “However, the danger of carbon monoxide poisoning from an internal combustion engine left running in an enclosed space has been well recognized for years in the automotive industry and by the public at large,” wrote Toyota Motor Sales’ Curtis Hamilton. “Unfortunate incidents such as the one that you describe in your letter, while rare, have also occurred with mechanical key systems and are not dependent on whether the vehicle involved uses a keyless ignition system.”

In March 2013, Zitser wrote back, observing that Toyota had missed her point entirely – an engine shutdown device could be implemented in any vehicle regardless of ignition type – and she encouraged Toyota to stand with safety:

“Again, I thought by sharing this tragedy with Toyota, it might take the opportunity to come up with a very simple solution that would prevent another accidental death – by installing an automatic shut off or kill switch. I hope you will reconsider this letter and not only be a leader in the automotive industry but equally important be in the forefront as far as the safety of the consumers of your product,” Zitser wrote.

By then, it was too late for Toyota to be the safety leader in using an automatic engine shutoff mechanism to prevent carbon monoxide deaths. In calendar 2012, both Ford and GM released MY 2013 keyless vehicles with automatic shut-off systems. And in 2017, Chrysler has added a similar feature to its 2018 Pacifica hybrid mini-van,

Zitser remains baffled that Toyota has refused to address the issue, given its long history of knowledge about the problem and the deadly consequences.

“I would never, ever, ever buy a Toyota,” she says. “The number of Toyota deaths in just this past year is staggering, and I just can’t believe they are not aware of what the other automakers have done to address this problem. I just don’t understand, first, why that the PARK IT bill hasn’t passed and second of all why Toyota has done nothing.”