For the last 100 years,
the gas burning engine has dominated and with the
engine come pollutants. Since 1971, the US
Environmental Protection Agency has been testing
these vehicles to ensure they comply with federal
pollution rules. But electric vehicles
made up 8% of all new car sales in the US in 2022,
a 55% increase over the previous year. This is the exciting time
right. Now here in the lab in the
industry. A lot of change going on,
a lot of new understanding. We're still kind of the
Wild West of these things. The regulatory
environment around electric vehicles is
still changing pretty fast. Evs don't have tailpipes
and they don't emit pollutants. But the
government still requires automakers to run
specific tests on their electric cars and publish
estimates of how far each can drive on a single
charge. And the agency tests some
of those vehicles itself to ensure the estimates
are accurate. The fear of running out
of battery called range anxiety is one of the top
concerns consumers have about EVs and something
keeping many of them from making the switch. The EPA's ratings are
considered more accurate for driving on American
roads than systems used elsewhere in the world. But the EPA system still
has its critics. One report found that EVs
fell short of their stated EPA range by more than
12%. They were less accurate
than the EPA fuel economy ratings for internal
combustion vehicles. In addition,
inconsistencies in testing rules can make
differences in range among brands seem larger on
paper than they are in real life. This issue where a Tesla
at with a 405 mile range and a Porsche with a 250
mile range are practically the same number in the
real world. So how is EV range
determined by the EPA? Why do some say the
ranges are not accurate and what can be done
about it? CNBC visited the EPA's
National Vehicle and Fuel Emissions Lab to find
out. We're here in Ann Arbor,
Michigan, at EPA's National Vehicle and Fuel
Emissions Lab. This is the federal
laboratory where all of the auto emissions and
fuel economy measurements are made. This car is an electric
vehicle, a Chevrolet bolt. It is about to undergo
one of the tests the EPA administers to determine
its range, how long it can drive on a single charge. We do more than car
testing here. We test engines from
large 18 wheel trucks. We test non-road
equipment that's in agricultural
construction. We test little engines and
lawnmowers and weed whackers and snow
blowers. Most of the time,
manufacturers test their own vehicles and report
the results to the EPA. It is similar to the
approach the National Highway Traffic Safety
Administration takes with safety certification. The EPA can test any car,
though it often will. If there is a substantial
vehicle refresh or update, which typically happens
every 3 to 7 years. They provide us with a car
that's either pre-production before it
goes into production or shortly after a job, one
when it's just into production. We actually
compare their submitted results with what we get
in the lab. Most of the time they
match up really good once in a while. They don't.
When they don't, then we figure out why. Usually sometimes we have
to retest. They have to retest. If their numbers in our
numbers don't line up, then they take EPA's
number as the official number that's generated
in this lab. So total 300 cars a year
get tested in the lab. That represents about 15
to 20% of the models that make up the US fleet in
any given year. 150 new cars and 150 used
ones to see how those are performing throughout
their entire life cycle. In addition to enforcing
its own rules and providing information to
the public, the EPA also passes data generated
here on to other agencies. Data goes to the
Department of Transportation for
corporate average fuel economy or Cafe
standards. The average fuel economy
an entire automaker's fleet gets. Numbers also go to the
IRS for the gas guzzler tax, a charge on all cars
that are below a certain mile per gallon rating. Originally there were two
tests a city test and a highway test. The city test is typical. A morning drive in Los
Angeles is actually where it was built on the drive
schedule built on the Los Angeles morning drive in
the mid 60s. The highway test, which
is more of a suburban test, was actually
derived from driving around here in the Ann
Arbor area. The EPA added three more
tests in 2008 to more closely match real world
driving conditions. A test in cold weather at
20°F. A hot weather test with
the air conditioner running and an aggressive
driving test. All the while,
powertrains were changing. The EPA began testing
hybrid vehicles once the Toyota Prius debuted in
the United States in 2001, it began testing electric
vehicles after it finalized rules for those
in 2010. It added equipment for
charging electric vehicles, for measuring
the discharge of electricity from the
battery on a battery electric or plug in
hybrid. But the dynamometers
themselves, the conditioning, the
temperature controls, the drive cycles, all of
those are the same, with just the addition of the
new features for the electrified part of the
vehicle. This is how the agency
tests an testers first charge the battery and
park the car overnight. The following morning it
is subjected to a series of tests on rollers
called Dynamometers or Dynos. These rollers are
meant to simulate at least some of the conditions of
driving. For example, engineers can add
resistance to simulate wind drag. Someone gets
into the driver's seat, the car is hooked up to a
machine that measures how much the battery is
discharging relative to the speed and distance
traveled. There is a city test, a highway test, and
a third test known as the steady state test, which
consists of running the vehicle at a constant
speed until the battery is fully depleted. Then testers charge the
battery again and time it. At first, the test was
simpler. The standard test was run
a city test Pause. Run a city test Pause. Run a city test. Pause
from when the battery was fully charged until the
car couldn't stay on the drive trace anymore. They ran a highway test
the same way as battery capacity has increased,
that has become prohibitive. We had to have three
shifts of drivers coming in and run these over and
over. So we went to a multi cycle tests and
those allow us to both combine city and highway
and to accelerate that depletion. So now, you
know, 8 to 10 or 12 hours we can deplete a battery
instead of needing to run those cycles over and
over again for 20, 24 or 30 hours. Part of the testing
process also involves applying a reduction
factor to the numbers A car gets in a test in
order to account for real world driving conditions. Lab results are
multiplied by a reduction factor of 0.7. Instead of the multi
cycle test, automakers can conduct a five cycle test
similar to the one used for gasoline vehicles. That version includes the
city and highway test, along with tests in hot
and cold weather and an aggressive driving test
at about 80mph. The EPA does do some real
world driving tests for fuel emissions and is
developing similar tests for EVs. So we are looking at how
the Dynamometer compares to some of the similar
travel that we would see out on the road and
comparing some of those things. But we also will
look at how a battery charging might happen out
on the real world. Figuring out how to test
for the effects of heating and cooling systems on
range, for example, is tough since there are so
many different options in any given vehicle for
heating or cooling a cabin. So whether it's a heated
seat or a heated steering wheel or a cabin heating,
all of these can have different effects on
range, determining which one is best to represent
the way the consumer is going to use that, and
then using that to provide additional information on
the effects of cold or hot is one of the areas that
we're doing work right now in the lab. That complexity is one of
the many challenges the EPA faces with EVs. We have a lot of
standards already for how combustion vehicles get
measured and the kinds of data that they need to
report to us in order to do certification and
compliance in order to drive on the roads. Those regulations are
quite, I have to say, wobble pretty obvious
yet, because just the kinds of data that
they're generating is just so new. We've had 100 years of of,
you know, looking at economy and efficiency
and emissions and everything on gasoline
cars we've only had at max ten years of it. And I
would say not even that really just a few years,
maybe 5 to 6 years where it's become important
enough that people are focusing on it. It isn't just the novelty,
it is also the pace at which everything is
happening. It is one of the fastest
changing areas that we deal with here in our
laboratory. Just in terms of how fast this
technology is moving. If you look at a vehicle
that we had in here even five years ago, a 2016 or
2017 electric vehicle looks almost completely
different to it than what we're seeing with
vehicles coming in 2024. So every manufacturer
kind of has their own way of reporting data on
where the power is coming into and going out of the
vehicle, how it's moving around between the motors
and the batteries, or if it's doing things like
regenerative braking, where the strategies
about how power goes to the heating and cooling
system. So versus how to take the battery at the
right temperature. All this makes it hard to
get lab results that match real world driving. I think that with our with
the addition of the more aggressive cycle and the
mid 90s, I think people are seeing that our fuel
economy is much closer to what you would see from
the values on the label versus what they're
seeing on the road with EV. It's a newer area,
right? Car and Driver magazine
tested a 2018 model three in cold conditions and
found it got only 65% of its stated range. Triple A tested electric
trucks under loads and found they lost 25% of
their EPA range. Edmonds, a site that
provides information on cars, found that many EVs
achieved ranges in their tests that were quite
different from EPA ratings. Some were quite
extreme. A Porsche Taycan has an
EPA estimate of 203 miles on a single charge, but
achieved 323 miles of range on a charge in
Edmunds tests. A paper published by the
Society of Automotive Engineers in April 2023
leveled several criticisms on how EV range is
tested. Among other things, the
paper said the EPA published only one total
range number for each vehicle, rather than
separating range ratings for city and highway
driving, as it does for fuel burning vehicles. That total range number
is weighted in favor of city driving over
highway. That is a problem, the authors
said. The paper drew on testing data gathered by
Car and Driver magazine. One of the paper's
authors, Dave Vanderwerp, is director of vehicle
testing at Car and Driver. In gas vehicles. The highway number is
almost always better than city. And in EVs it's the
reverse. Right? So people aren't
seeing that, Oh, the highway is quite a bit
worse than city when you just get this one
combined number. Evs are often equipped
with regenerative braking. That is a mechanism that
captures energy from the wheels as the vehicle is
slowing down and stores it back in the battery. This gives them an edge
in city driving where there is a lot of
speeding up, slowing down and stopping. This is
precisely where gas powered vehicles are less
efficient. Even the EPA highway test
involves a lot of speeding up and slowing down,
which gives regenerative braking a chance to kick
in. Let's say you had 400in
the city. And 300 on the highway. That gets combined to
something like 340, and that's what shows up on
the label. However, I think most
people will have range anxiety concerns when
they're on the highways. One car and driver test
consists of driving a vehicle at a constant
75mph. Most highways in the US have speed limits
of 70mph, with a few states allowing as high
as 85. Car and driver testing
data found that while internal combustion cars
actually outperformed their EPA highway ratings
by about 4% on average, EVs fell short of their
range number by 12.5%. It shows that range can
vary wildly depending on how you drive the car. If you're driving in one
part of the country under one set of conditions
where it's really, really hilly, or if you're
driving somewhere else where it's very flat, if
you're driving in a place where it's cold or it's
really hot, this one number will allow you to
compare from model to model, but it might not
represent what you'd actually get in all of
those different driving cycles. The paper's authors,
Pinone and Vanderwerp, even say putting the
total range on the Monroney label is not
accurate to real world use. No one drives the
car down to zero miles. A car might have a stated
range of 300 miles, but the vast majority of
drivers are likely to recharge before the
battery runs too low. Different brands often
report very different total ranges on their
vehicles. But Pinone and Vanderwerp
say the differences might not amount to as much in
real life and may be at least partially
attributed to the fact that the EPA allows them
to use different methods to report the range on
their cars. Either the Multi-cycle
test made up of a city and highway test or a five
cycle test that incorporates hot weather,
cold weather and aggressive driving. About 28% of gas burning
vehicles use the five cycle test, but about 50%
of EVs do. Why is that? Because it's
beneficial tip In the gas world, the five cycle was
generally deemed as not beneficial. So the
automakers kind of stayed away from it if they
could, because a lot of times come up with a
worse number. But Pinone and Vanderwerp
say it can give an EV a higher EPA range number
than the standard multi cycle test with its 0.7
reduction factor. Their paper recommends,
among other things, that the EPA publish both city
and highway range apply a single multiplier close
to 0.6 to get testing closer to real world
circumstances and make every automaker run the
same tests. I think the range testing
right now is relatively stable. It's pretty
representative. It's not obvious that
there's streamlining that can be done that would
significantly reduce the test time that it takes
to run these tests and still get representative
numbers. But that work is being
examined and studied at by different manufacturers
and by some independent labs. The EPA said the Range and
fuel Economy section of the Monroney sticker
hasn't been updated since 2011, when there were few
electric cars and plug in hybrids. At the time, the
decision was made to combine the city and
highway results to keep it simple for consumers. But the agency is
continuing to evaluate the procedures and methods
used for determining EV range and labeling, as it
has done for gas vehicles throughout its history. It is revising the cold
test, for example, for the 2025 model year based on
Tesla's 2023 results. The EPA estimates that
for the Tesla Model Y performance, the new
range estimate would theoretically decrease
from 303 miles to 289 miles due to changes in
the cold test, It said it is reviewing the recent
SAE paper, but that it would be premature for us
to comment on any specific ideas or observations. Automakers have a long
history of pushing against regulations, trying to
find loopholes and sometimes just lying and
defrauding the public. The Dieselgate scandal is
perhaps the most recent infamous example of this. A number of automakers,
but most notoriously Volkswagen, were caught
using devices intended to cheat on EPA emissions
tests. The scandal rocked VW,
currently the world's second largest automaker,
leading to billions in fines and payouts and
jail sentences for top executives. It also
precipitated the German carmakers dramatic pivot
toward electric vehicles and helped fund VW nearly
$2.5 billion Electrify America Charging
Infrastructure Project. The scandal also shaped
the EPA. They simply don't have the
resources to to look hard at electric vehicles. In fact, they are still
reeling from the Dieselgate and and trying
to ensure that that, you know, from an emissions
standpoint that we're doing the right thing for
the environment and can 100% appreciate that. Right. Electric vehicles
really are not bad for the environment in the same
way that gasoline powered vehicles are. There are still concerns
consumers need to be protected. If automakers have been
sort of allowed to stretch the regulations further
than maybe they should. Getting that back in the
bag is going to be coming increasingly harder. And now with the
governmental push towards EVs. It's time to. It's time to make sure
that. That it's right. Every test we do for is
just a little bit more standing on the whole all
of what can what your was do, what might they do in
the future and what what should we look out for
when the next vehicle comes it. We're consumers as well
and you know, scientists and engineers looking at
this stuff. So it's really important
for us to look at what might those issues be and
how do we apply it and how do we test for those
things as well.
