Revolution Requires Evolution: We Need New Roadside Safety Standards for Electric Vehicles

In June 2024, TTI's Roadside Safety and Physical Security Team crashed a Tesla Model 3 electric vehicle (EV) into a heavy-duty guardrail at 62 miles per hour. When the EV blew right through the barrier, researchers were stunned. TTI Senior Research Engineer Roger Bligh, whose 38 years of roadside safety barrier testing experience oversaw the test, joins guest host and TTI Agency Director Greg Winfree to discuss the results of the testing and the broader implications for standards governing the development and deployment of roadside safety devices. | View the Crash Test

Episode Transcript

Greg Winfree (Guest Host) (00:14):
Hey everyone. Welcome to Thinking
Transportation--conversationsabout how we get ourselves and
the stuff we need from oneplace to another. I'm Greg
Winfree, agency director of theTexas A&M Transportation
Institute, and I'll be yourguest host for this episode of
Thinking Transportation. A bitfewer than 1 percent of total

(00:35):
vehicle registrations in Texasare electric vehicles, but
almost 7 percent of newvehicles sold in Texas last
year were EVs. So the number ofEVs on Texas roads will
continue to grow over time.
Ensuring that these EVs operatesafely and collaboratively with
other road users will be afunction of how well our
roadside safety standards matchup with these new types of

(00:58):
vehicles. Today, I'll betalking about that with Dr.
Roger Bligh, TTI's seniorresearch engineer and Roadside
Safety Program Manager. Rogerhas a PhD in civil engineering,
is a licensed ProfessionalEngineer, and was named a
Regents Fellow by The Texas A&MUniversity System Board of

(01:19):
Regents in 2010. Roger, thanksfor joining me.

Roger Bligh (01:23):
Hi, Greg. It's a pleasure to be with you today.
I always enjoy opportunities todiscuss our research program
and roadside safety.

Greg Winfree (Guest Hos (01:31):
That's awesome. And I believe our
listeners are really gonna beenergized about hearing this
conversation. So why don't westart out by level setting .
You know, it seems obvious fromthe name, but what exactly does
TTI's Roadside Safety Programfocus on?

Roger Bligh (01:47):
So when we use the term roadside safety hardware,
that generally refers to anydevice that is intended to help
reduce the severity of aroadside crash. So we're
talking about devices such asguardrail or median barrier,
like concrete barrier or cablebarrier, bridge rails, crash
cushions, all those types ofdevices. And our program

(02:11):
designs tests and evaluatesthose devices to help improve
the safety of motorists shouldthey happen to depart the
roadway.

Greg Winfree (Guest Host (02:20):
Okay.
I know folks are familiar withguardrails and cable-medium
barriers. They're the kind ofequipment that are on roadways
across the country, but theynever really focus on those. So
let's talk about the safetystandards. How are the safety
standards set for guardrailsand barriers?

Roger Bligh (02:37):
Well, the testing standard itself is an AASHTO
document, the AmericanAssociation of State Highway
and Transportation Officials,and it's called MASH--Manual
for Assessing Safety Hardware.
So it's that document thatrecommends the testing and
evaluation criteria forbarriers or other types of
roadside safety hardware. Andthe impact conditions in that

(03:01):
standard are developed fromreal-world crash data. So we
periodically review that dataand develop real-world
distributions, and we use thatdata to understand how we
should be testing our devices,what speeds and angles we
should be using. And in termsof the vehicles, we also
periodically review sales datato determine what types of

(03:22):
vehicles we should be using fortesting.

Greg Winfree (Guest Host (03:24):
Well, MASH is obviously a pivotal
document in this space, but dostates have the authority to
develop their own safetystandards? Or is this a federal
responsibility, or is it acombination of experts in these
areas dictating what goes onour roadways?

Roger Bligh (03:40):
Yes. Well, the Federal Highway Administration
has adopted the use of MASH onthe National Highway System.
And so the states will developtheir standards for use on the
National Highway Systemfollowing the MASH criteria. So
they develop their ownstandards and can adopt
hardware from each other asdesired to meet the specific

(04:02):
needs that might exist on theirhighway network.

Greg Winfree (Guest Host) (04:05):
And how can state DOTs be sure that
manufacturers of roadsidesafety hardware have products
that meet the standards ?

Roger Bligh (04:14):
Well, this MASH standard that we've been
talking about is a performancestandard. So the roadside
safety devices that are beingdeveloped are actually
physically tested to theseprescribed test matrices to
help ensure that they'reperforming in a certain way and
will provide the level ofsafety for motorists that's
desired. So, for example, if wewere using a barrier, it's

(04:36):
expected that that barrier willcontain and redirect an
impacting vehicle within thetest conditions. And what I
mean by that is that thevehicle should not go through
or over or under the barrier.
So the objective is to keep amotorist from striking an
object or encountering somenon-reversible terrain on the

(04:57):
roadside. And the barrier isgoing to be recommended when
the severity of impacting thebarrier is less than what it
would be if you were strikingthat roadside obstacle.

Greg Winfree (Guest (05:09):
Understood . Well, as you know, when
visitors come to the TTIfacilities here on the RELLIS
Campus, the first thing theyask to see is a crash test.
Now, you and your team are someof the most popular folks here
at TTI. How long have youpersonally been involved with
crash testing and how long hasTTI been conducting crash-test
research over the years?

Roger Bligh (05:30):
Well, it's hard to believe it's been this long,
but , I've beendesigning and testing roadside
safety devices for 38 yearsnow, and TTI has been
performing crash tests ofroadside hardware since the
1960s. It's always a pleasureto host visitors for a crash
test and talk to them about oursafety research and our program

(05:51):
because you know, roadwaydeparture crashes, they're the
leading cause of crashfatalities in the U. S.,
representing over 50 percent ofour roadway fatalities. So it's
a really significant problemwe're trying to address, and I
think the more information weget out about that, the better.

Greg Winfree (Gues (06:08):
Absolutely.
And you know, witnessing acrash test , folks really come
to understand the importance ofthese technologies. And like
you said, the fact that thesecrash scenarios are the leading
cause of fatalities is not loston any of the visitors that
come to TTI. But interestingly,when I joined TTI in 2016, I

(06:28):
was completely blown away bythe level of science,
engineering, and technologyrequired to ensure the safety,
repeatability, and consistencyof crash tests. And you break
down for our listeners thepainstaking detail necessary to
conduct crash tests. I mean,how do the vehicles get to
certain speeds and certainangles? And in addition to

(06:49):
learning whether roadsidesafety hardware performs as
expected, what else do youmeasure and monitor?

Roger Bligh (06:56):
Yeah, those are good questions. Let's consider
a barrier system as an examplefor discussion. After the
design of that barrier has beencompleted, we construct a
prototype of the barrier at ourproving grounds and then a
vehicle of some prescribed bodystyle, you know, according to
our standard is purchased andprepped and instrumented. And

(07:17):
then what we do is, we use atow vehicle. And the tow
vehicle is used to pull thetest vehicle up to the
designated impact speed using atow cable and some pulleys and
a separate guide cable, whichis tensioned up along the path
of the vehicle. It feedsthrough a guide arm attached to
the front wheel of the testvehicle, and that serves as a

(07:37):
form of external steering, andit helps guide the vehicle into
the barrier at the desiredlocation and angle. And then
both of those cable systems,the tow cable and the guide
cable, they release from thevehicle prior to impact, and it
leaves that vehicleunrestrained and free to react
with the barrier system.

Greg Winfree (Guest Host) (07:55):
So I've seen several tests of
course, and you utilizedifferent angles and
velocities. What's the sciencebehind that or the rationale?
How were they selected, andwhat are the range of
performance parameters thatneed to be passed for a system
to be successful?

Roger Bligh (08:14):
Yeah , so the standard does prescribe certain
speeds and angles that we useto evaluate the hardware. So
it's a performance standard,and we run those tests to a
prescribed matrix. And some ofthe details of those tests such
as the speed and the angle thatwe use for our barrier systems,
they actually originate with areview of real-world crash

(08:34):
data. So we periodically willlook at that crash data and
reconstruct some of thosecrashes to understand what the
impact speeds and angles are.
And from that information,we're able to develop a
distribution for impact speedand impact angle and then
select a percentile that wefeel is appropriate for design

(08:55):
purposes, and that informationgets fed into our standard.

Greg Winfree (Guest Host (09:00):
Okay.
Now, we didn't talk asignificant amount about some
of the pre-work that's done inthe Center for Computational
Mechanics. Maybe a little bitabout how you all use
simulation to suss out testseven before the physical test
is conducted.

Roger Bligh (09:16):
Yes. So in fact, I mentioned that part of what our
program does is we're not justtesting these barriers, but
we're also designing them. Andthe computer simulation is a
tremendous tool that we haveavailable to us to help in that
design process. So what we cando is we can actually model the
roadside safety hardware systemthat we're evaluating if it's a

(09:38):
barrier or something else. Andwe also have finite element
models of these test vehicles.
And so we're able to, on thecomputer, before we ever go out
to our test track, we're ableto help determine the predicted
impact performance of aparticular system. And that
helps us to perhaps make somedesign changes before we get to

(10:00):
the test track or even optimizethe design. And so the computer
simulation has become anextremely valuable tool for us
in the overall process ofdesigning hardware and getting
it out onto our roadways.

Greg Winfree (Guest Hos (10:12):
Right.
Well in addition to braggingabout the prowess of your team
and the work that you do, whydon't you talk a little bit
about the scanning and thedisassembly of vehicles so that
you get all of the parametersneeded so that the computer
simulation is successful, whichthen means the physical test is

(10:33):
more likely to be successful?

Roger Bligh (10:35):
Right. So that goes to that modeling process
that I was mentioning and thefact that we need models of our
design vehicles right for usein these computer simulations.
And so those vehicle models aretypically developed through a
reverse-engineering process. Webasically disassemble the
entire vehicle and weessentially scan in all of

(10:56):
these different components andthen eventually put them back
together. And so we're lookingat all of the key features of a
vehicle and even the connectionbetween these parts such as
where the spot welds are andthe different types of
connections. And we alsodevelop material properties
through some physical coupontesting. And so all of that
goes into the construction ordevelopment of a vehicle model

(11:20):
that we can use to help assessthe performance of our hardware
through the development of thehardware models that I
mentioned.

Greg Winfree (Guest Host) (11:28):
Now, that's just tremendous work and
that's why I wanted to drilldown so that our listeners
really understand how complexthis undertaking is. So I'd
like to reflect back on a crashtest from last June. And for
our listeners, we will beattaching a link to the video
of the crash test that we'rediscussing today in the show
notes. Your team tested an EVagainst the heavy duty

(11:50):
guardrail. Now what were thereasons behind selecting that
particular type of roadsidesafety hardware?

Roger Bligh (11:57):
Yes. Well, prior to that test that we performed
some of our colleagues at theMidwest Roadside Safety
Facility at the University ofNebraska, they had performed
two crash tests on a standard,what we refer to as W-beam
guardrail system using electricvehicles. And so the W-beam
guardrail, that's the mostcommonly used guardrail system

(12:18):
across the country, it's whatyou're probably seeing on your
roadside is a metal beam onsome posts. And it's been shown
to be MASH compliant throughseparate full-scale crash
testing. So the EV tests of theW-beam guardrail, they were
performed with a rivian R1Tpickup truck and a Tesla Model
3 passenger car. And both ofthose tests were performed at a

(12:43):
nominal speed of 62 miles anhour and an angle of 25
degrees, which are the impactconditions that correspond to
the MASH document that Imentioned for what we call test
level three, which is the basictest level defined for
passenger vehicles on our highspeed roadways. Well, the
Rivian pickup truck actuallyruptured and went right through

(13:06):
the W-beam guardrail system.
And then the Tesla, which is asmaller vehicle, was actually a
pretty interesting test. TheTesla Model 3 actually
underrode the W-beam guardrail.
So we had two failures of theW-beam guardrail system that
kind of informed us prior to usperforming our test.

Greg Winfree (Guest Hos (13:26):
That's very eye-opening. So if
previous tests demonstratedthat a standard W-beam
guardrail couldn't contain andredirect the EV, I mentioned
that it was a heavy-dutyguardrail, but specifically
what kind of guardrail did youtest this summer?

Roger Bligh (13:40):
Right. Well, given the failure of the W-beam TTI
decided to test, as youmentioned, a stronger, more
robust guardrail system. Andthat system is referred to as a
thrie beam guardrail. And thethrie beam rail element is
deeper compared to the W-beam,right. And so for instance, the
thrie beam is 20 inches deep,whereas the W-beam is 12 inches

(14:04):
deep, it actually had theeffect of reducing the ground
clearance, the height from theground to the bottom of the
rail. And so we thought thatwould be helpful in addressing
that underride behavior that weobserved in the W-beam
guardrail test with the Tesla.
And so the deeper thrie beam,it also has a greater
cross-sectional area, so thatmeans that it's stronger than

(14:24):
the W-beam. So this thrie beamwas also successfully crash
tested under MASH criteria, andit was considered the next
level of guardrail to evaluatewith EVs.

Greg Winfree (Guest Ho (14:35):
Mm-hmm.
So consistent with whatNebraska-Lincoln utilized--the
vehicle they used -- it soundslike we also used a Tesla Model
3 so that we could compareapples to apples. But what were
the results of the tests thissummer?

Roger Bligh (14:48):
So you are absolutely correct. We did use
a Tesla Model 3 in thisparticular test, and we used
the same impact conditions. Andit was to be able to assess how
this thrie beam guardrail wouldperform under those test level
three conditions, you know,with that particular vehicle.
Well, during that test with theTesla, the vehicle kind of

(15:09):
began to wedge under the thriebeam rail. It kind of
compressed the bottom edge ofthe rail upward and eventually
ruptured the rail and wentthrough the thrie beam
guardrail. And this was reallyunexpected because the impact
severity of the Tesla waswithin the range of the current
MASH testing criteria. Mm-hmm . So we didn't

(15:31):
really view this as a strengthtest with the pickup truck. We
were trying to evaluate thatunderride behavior with the
smaller Tesla car. And so oneof the reasons that surprised
us is because we had seen thethrie beam work with other
types of heavier vehicles overthe years, including a single
unit truck and even a bus. AndI'd never seen a thrie beam

(15:52):
guardrail rupture during my 38year career.

Greg Winfree (Guest Host) (15:56):
So what in particular about the
EV--or is it specific to theTesla?--but what might explain
those results? I mean, how doesthe weight, center of gravity
materials, and constructioncompare to standard sedans with
internal combustion engines?

Roger Bligh (16:11):
Oh , that, that's a great question. I think it's
very important to point out,this is not a Tesla problem.
This is an EV problem, right?
And we know there are manycharacteristics of EVs that are
different from their internalcombustion engine (or ICE)
counterparts. And there arecertain key differences that
affect the compatibility of EVswith roadside barriers. So you

(16:35):
mentioned that the EVs areconsiderably heavier than their
ICE counterparts. You know, forexample, a Ford Lightning EV
pickup weighs 2,000 pounds morethan the conventional Ford
F-150 pickup.

Greg Winfree (Guest Host) (16:46):
Wow.

Roger Bligh (16:47):
And that increase in weight, it places a lot more
demand on our barrier systems.
And so a barrier that might benear its performance limits
under the MASH standard, it mayfail when impacted by the
heavier EVs. Mm-hmm. Mm-hmm
. And you alsomentioned, Greg, the center of
gravity height. And that'sanother key difference. So the
weight and position of thebatteries in these EVs, right,

(17:10):
it lowers the center of gravityheight compared to a
traditional ICE vehicle. Thinkabout these batteries forming
kind of a skateboard type ofarrangement underneath the
occupant compartment. Well,although that can help
stabilize a vehicle compared toa normal ice counterpart, it
can also increase the potentialfor that vehicle underride. And

(17:30):
it's that type of underridebehavior that we saw with the
guardrail test. And it was alsoa partial factor in the thrie
beam failure as well. Andthere's some other properties
that can contribute to theseundesirable test results that
we've seen. And those includethe crush stiffness of the EV
and the profile of the EV. Sowe don't have a conventional

(17:51):
engine compartment and there'sno front engine. So the crush
profile in the front quarterpanel area that interacts with
the barrier, it's considerablydifferent. And that can permit
the guardrail elements tointrude further into the
occupant compartment andinteract with other components
such as the wheels that canmaybe initiate tears in the

(18:11):
guardrail.

Greg Winfree (Guest Host) (18:12):
Mm- hmm . Well,
knowing that Tesla's sisterorganization is SpaceX, are
they also utilizing space-agematerials or anything else
exotic that might even beanother factor for
consideration?

Roger Bligh (18:25):
Well, certainly can be. The design of those
vehicles is different from ageometric standpoint. The roofs
of those vehicles, they tend tobe all glass. So there are some
other differences compared toour ICE vehicles. Certainly the
battery compartments downthere, really rigidize parts of
the frame. So there's a lot ofother smaller differences that

(18:47):
can also contribute to howthese vehicles interact with
our roadside safety hardware.
And we really need to get abetter grasp on that and start
looking at how we can maybeimprove that interaction.

Greg Winfree (Gues (18:59):
Understood.
Well, you know, I did somequick online research and
discovered that there werealmost 250,000 EVs on Texas
roadways and an estimated 3.3million EVs on the roads
nationwide. So what are some ofthe implications of these
results as more EVs are takingto the road across the country?

Roger Bligh (19:18):
Right. We know these vehicles are increasing
in number. It's not a matter ofif this is going to occur, it's
occurring now. Right. And asthese numbers of EVs continue
to grow, the exposure in termsof the number of vehicle miles
traveled also increases. Andthis is going to lead to an
increase eventually in thenumber of crashes, including

(19:40):
crashes with roadside safetybarriers. And so if this EV
compatibility with barriersystems is not addressed, it
can result in an increase inthe number of serious injury or
fatal crashes that we might seewith barrier systems in the
future.

Greg Winfree (Guest Host (19:57):
Well, let me ask another question. So
what happens next from adevelopment perspective in
terms of different vehicletests or new kinds of hardware
that can sustain impacts fromincreasing vehicle weights?

Roger Bligh (20:09):
Right. Well, our current MASH standard that
we've been talking about today,it does not address EVs yet. It
takes time for these standardsto be updated to reflect
changes in vehicle technologiesand other areas. But that
doesn't mean that we shouldn'tbe looking at this problem now.
Right? We have identified thatthere is a compatibility

(20:30):
problem between EVs and some ofour existing roadside barrier
systems. And we know that thenumber of EVs in our roadways
is going to continue to grow.
So what we need are new barriersolutions that are compatible
with EVs. We need to developand test and implement those
barriers to help address thissafety issue that we've
identified. And I should notethat the safety hardware for

(20:54):
EVs, as we develop that, Ithink it's also going to have
an added benefit in terms ofaccommodating our heavier
conventional engine vehicles,our ICE vehicles, that are
currently not addressed by ourtesting standard. Because
there's a general trend, notjust with EVs, but also with
our ICE vehicles of increasingmass, increasing weight. And so

(21:16):
as we start addressing the EVproblem that we've identified
and implementing barriersolutions for EVs, we'll have
an added benefit in that areaas well.

Greg Winfree (Guest Host (21:27):
Okay.
Well, Roger , you've been doingthis kind of work for a long
time, but it's clear to me thatyou're still energized by it.
What are some of the reasonsyou're ready to come into work
each day?

Roger Bligh (21:37):
Right, well, my biggest motivation is the
safety aspect of my work. And Ifeel like most of us have been
impacted by a vehicle crash insome way. I know that I have. I
often think about how somethingI do today could possibly save
a life in the future, andthat's very empowering. So I
always say, you know, hopefullymost of us will not have to

(22:00):
rely on a roadside barrier or asafety device in our lives, but
if you do, I'm gonna do my bestto try and ensure that it's
there and ready for its job ifyou need it.

Greg Winfree (Guest Host (22:11):
Well, those are tremendous parting
words and truly amission-oriented approach that
is greatly appreciated. So Dr.
Roger Bligh, thanks for joiningme to talk about the important
work TTI is doing to keep upwith the never-ending changes
in the vehicle fleet here inTexas and around the nation.

Roger Bligh (22:30):
Well , it's been a pleasure, Greg, and I really
appreciate you having me. Thankyou.

Greg Winfree (Guest Host) (22:35):
So folks, in addition to exploring
the safety implications of moreelectric vehicles on our
roadways, researchers at TTIare also engaged with sponsors
to determine how EVs can reducetransportation sector emissions
and related health effects, andhow EV adoption will affect our
electric grid and ourtransportation funding methods.

(22:57):
Although the majority of roadusers don't realize it, they
owe a huge debt of gratitude tothe world-class researchers at
TTI and other institutionswhose mission is to ensure that
the road-going public gets homesafely. Thanks for listening.
Please take just a minute togive us a review, subscribe and
share this episode. I inviteyou to join us next time for

(23:19):
another conversation aboutgetting ourselves and the stuff
we need from point A to pointB. Thinking Transportation is a
production of the Texas A&MTransportation Institute, a
member of The Texas A&MUniversity System. The show is
edited and produced by ChrisPourteau. This is Greg Winfree,
signing off. Thanks again forjoining us. We'll see you down

(23:41):
the road.