How the car industry hid the truth about diesel emissions

https://www.theguardian.com/environment/2019/mar/22/dirty-lies-how-the-car-industry-hid-the-truth-about-diesel-emissions

by NeedMoreTea • 7 years ago

170 148 7 years ago

theguardian.com

joelkevinjones• 7 years ago

A coworker and I had a discussion with another coworker when this scandal broke. We couldn't convince him that he was wrong when he said there is no way car software could "cheat" like that. Even after we came up with several things that could be done in the space of about a minute, he remained unconvinced. If you follow through the article to the reference US DoJ press release, VW's software used several of the ideas we came up with as part of their "cheat". This coworker also thought that the California drought was a big hoax, so...

oska• 7 years ago

What's the point of this anecdote other than to make you look smart and your coworker look dumb? Maybe both things are true but it's hardly a very interesting observation that there are both smarter and dumber people in the population.

Gibbon1• 7 years ago

You know I wouldn't have have read his comment if you hadn't wrote this one.

gcb0• 7 years ago

in the middle he points how the software "features" were decidedly malicious as they are similar to their own hypothetical solutions to circumvent the law?

Theodores• 7 years ago

It is how scandals work. Lots of people believe the lies and can't be persuaded to think from an evidence first viewpoint. They need the story to be headline news on the BBC or some other reputable source before they will 'believe it'.

Every scandal relies on groupthink to keep it quiet.

godelmachine• 7 years ago

Excellent review by Adrian Colyer for those interested - https://blog.acolyer.org/2017/06/20/how-they-did-it-an-analy...

Also, the book How they did it gives insightful glimpses.

makomk4• 7 years ago

" Schmied responded that as long as emissions went down when limits were tightened, his department didn’t mind they were many times higher than allowed."

Some context for this: the official European test cycle was known not to be representative of real driving conditions. Because it was less aggressive than real-world driving, it pretty much always overstated fuel efficiency and understated emissions. So real-world emissions not going down at the same rate as tested emissions is a sign of foul play, but them being higher isn't necessarily. (The US test cycle is better.)

mirimir• 7 years ago

Sure, but don't the laws limit emissions? Irrespective of how well standard government tests were measuring those emissions?

So "We publish this data" just sounds like "I'm doing my job. Anything more is not my problem."

makomk4• 7 years ago

The laws limited emissions in the standard government test cycle, and the thresholds on emissions were chosen based on how well cars could be made to perform in those test cycles. While actual defeat devices were illegal, the legal emissions limit wasn't actually the real-world emissions limit and everyone knew it.

(The EU is now gradually moving to real-world emissions testing and a more realistic test cycle, though judging from the other reply they might have some cheating problems with that - especially as real-world testing inherently has wider errors than lab testing.)

rurban• 7 years ago

The US test cycle is better.

Then you never heard about the US test cycle cheating scandal. It's not better, it's actually simplier to cheat. The US test cycle is tailored to a more realistic city environment with a lot of stop's and go's, cold starts, and requires an actual professional driver. No automation. These drivers learned soo much how to cheat the system, that they were in high demand and they were all held exclusively by the US car industry, so Asian and European companies were at disadvantage. That's why the Asian industry agreed to use the European test cycles and not the more realistic US system. It was too expensive and too easily tricked. The European standard was cheaper and fairer (those times). Until they found a way to trick it also.

ummonk2• 7 years ago

I remember Mazda having trouble making their diesel engines performant while conforming to California emissions regulations, and wondering why they were having so much trouble compared to European car manufacturers. Then this scandal broke and it was obvious why.

alkonaut3• 7 years ago

I wonder why mazda or some other competitor didn’t find this by doing their own research? I make software for structural engineering and when some competitor launches a program that makes “better designs” it has usually turned out to be because they simply don’t follow the design codes/regulations. We wouldn’t be pushed from the market by that - we’d fund the analysis, reverse engineering and legal processes necessary to show the competitor is cheating.

sjwright• 7 years ago

If Mazda was attempting and failing to get their cars to pass the defined emissions test, one can imagine they were only ever testing their own and their competitor's cars under those test conditions—which we now know triggered the cheat mechanism in VW cars.

But yes, had the research group not performed their on-road tests, it would have only been a matter of time before an honest car maker would have realised their competitor's sums don't add up and perform or fund their own independent investigation.

(Perhaps this is exactly what happened behind the scenes? It's known that CARB ordered the research due to discrepancies in US v EU numbers—but someone pushed them to look at the numbers closely. They put a lot of work in to perform these tests so it couldn't have been on a whim.)

bboreham• 7 years ago

My conclusion, from the first time I thought about it, is that they were all cheating.

Engineers move job; they talk to each other. Every car company had to have people who knew the truth and, if they were being unfairly competed against, could publish this truth.

blub• 7 years ago

Were they having technical trouble or just having trouble finding a cost-efficient solution?

AFAIK the technical solution (Urea injection) works fine, but it's expensive and VW didn't want to include it in all cars.

vvillena• 7 years ago

VW uses urea injections, but they wanted to make it transparent to the end user. The urea was to be refilled during maintenance, and disconnecting the system was the only way to make the tank last that long.

TheRealDunkirk5• 7 years ago

I work in this industry, so I've been following it from the start. I've run test cells and written software to analyze the results from the emissions cycles that the government uses to validate the emissions compliance. I have worked around and for a very large diesel engine manufacturer who I am convinced does not cheat on this test. You'll understand why, below.

I recently leveled-up my career, and I now work with the person who originally setup such testing at this manufacturer, and one of the two people who literally wrote the spec, which is one of the most thorough technical documents I've ever laid eyes on. These are 2 of the highest-caliber engineers I have ever met in my 25-year professional career, both in terms of intelligence and capability, and in terms of integrity and trustworthiness.

This quote from the article pisses me off: "What I understand now is that the people we entrusted with the power to protect us essentially decided not to bother. Instead, they have allowed carmakers to spew whatever they want into our air." Really?! The government commissioned the spec. The government mandated that engine makers make their engines comply, and validate this compliance with internal testing. The government then paid independent testers to validate the manufacturer's results.

Excuse me, but what more did the author expect the government to do? The regulators entrusted this process to expert engineers, and expert testers. Did the author expect people who could invent the spec and create a test cell to do the validation to go to work for the EPA directly, and do the paperwork, instead of the engineering? Did they expect them to escrow the code, and validate that there was no cheating? Possible, but, again, you're asking a very specific skillset to give up DOING, and instead REGULATING, and that doesn't usually work.

The fault of this scandal lies -- of course -- with the manufacturers, but the fault also lies with the legislators. To control diesel engine emissions within the spec requires an array of very expensive after-treatment components. That can work in the class-8 heavy trucking industry, but you can't even fit it into a passenger car, let alone make it affordable at gasoline car prices. To think that cars could use diesel engines, be cost-competitive with gasoline-fueled alternatives, AND pass the ridiculously-low emissions standards that lawmakers have arbitrarily decreed was simply wishful thinking.

ip262• 7 years ago

I guess it depends a little on whether you feel like the spec writer or testers should have anticipated cheating, and attempted to detect it or not. If the job of a tester includes catching cheaters, or if a good spec should be very hard to cheat, I'm not sure missing that falls at the feet of the legislators.

wil4212• 7 years ago

How can VW create an engine, cheat, and sell it to Chrysler? Then Chrysler has to put a urea tank to meet emissions. How did some brilliant engineer writing the spec or the third parties not notice?

TheRealDunkirk5• 7 years ago

An SCR (and EGR, and DOC, et. al.) is a "given" in this equation. There's no hope of passing emissions without it. Even on gas engines, you still need a catalytic converter to pass.

ummonk2• 7 years ago

To think that cars could use diesel engines, be cost-competitive with gasoline-fueled alternatives, AND pass the ridiculously-low emissions standards that lawmakers have arbitrarily decreed was simply wishful thinking.

The emissions standards are perfectly sensible for maintaining good air quality. It definitely looks like it is wishful thinking for diesel engines to meet those standards and be cost-competitive with gasoline-fueled alternatives, but there is nothing wrong with that. Gasoline engines may simply be the more suitable technology for passenger cars.

Reason0772• 7 years ago

It definitely looks like it is wishful thinking for diesel engines to meet those standards and be cost-competitive with gasoline-fueled alternatives

The technology that would have brought "Dieselgate" diesels into compliance with EU emissions rules was said to cost manufacturers something on the order of US$100 per vehicle: Selective catalytic reduction (SCR) with a refillable AdBlue (urea) tank.

However, prior to Dieselgate there is alleged to have been collusion between the major German automakers and a secret agreement not to install it on small vehicles.

This collusion, rather than the emissions cheating itself, is the focus of ongoing criminal investigations in Germany. (Cheating on emissions tests itself was not actually illegal in Europe, because the legislation was created with loopholes).

As of 2019, AdBlue is becoming common on new small diesel vehicles in Europe.

topspin2• 7 years ago

"Excuse me, but what more did the author expect the government to do?"

It is not some outrageous expectation that the $8,000,000,000+/year, 14,000+ employee EPA and/or DOT independently verify the real world emissions performance of passenger cars and other vehicles. The cars were so grossly out of compliance that anyone with a ordinary field emissions testing machine could have caught it.

If, as is very likely the case, the regulators had no technical resources among their hordes of lawyers they could have asked any number of publicly funded universities to perform the actual investigation and go beyond the agreed upon and easily fooled test regime. The discovery was actually made by US academics with very limited financing.

This institutional pencil whipping is SOP at EPA. GOA had a gasoline powered alarm clock (among other improbable devices) Energy Star certified as part of an investigation in 2010. Energy Star certifications are literally pencil whipped through an automated system with no human review.

When the EPA found itself answering for their part in the emissions scandal in public hearings there was no defense. They didn't even try to cite limited budgets[1] -- the go-to excuse every time some regulator comes up short -- as a cause. They just failed.

That's difficult to accept for some, and so all manner of rationalizations are employed to dilute that simple truth. But I'm not among them.

[1] https://www.c-span.org/video/?c4787386/blaming-budget

TheRealDunkirk5• 7 years ago

I still shake my head at this. If you wrote laws that said manufacturers of a particular thing had to do X, and expected them to adhere to it internally, and then paid expert testing facilities to validate those results, why would you EXPECT that you would have to do additional testing to validate the validation?

And further more, why would you think that merely testing the test would lead to a different result? You can say that the Euro V is "not representative" (as someone else did in this thread) of normal driving, but the FTP test's torque and RPM curves were lifted from real-world driving through hilly areas. It most definitely is representative. I don't know about the Euro V.

So I don't get it. In this scenario, what people are asking the EPA to have done was load up vehicles with mobile emissions equipment (which is a very difficult thing to do, by the way), and validate that the TEST ITSELF correlated with real-world driving. Granted, this is an interesting exercise, but one that I would expect the EPA to - again - pay someone else to do.

Given the engineering that went into the spec on the US side, I wouldn't expect that this was something they should have needed to do. You can say that they were the regulatory body in charge of this, and therefore should be responsible for catching cheaters, but I don't see how a reasonable person can find that the EPA was intentionally negligent in this matter.

specialist• 7 years ago

We need better models of governance.

One party arrangements (self-policing) ain't working.

Two party arrangements (regulation) is hit or miss.

Maybe add a third chair to each negotiation. Just like the US Constitution has three branches of government. What would that look like?

Maybe incentivize competitors to narc on each other. What would it have taken for Mazda (for example) to blow the whistle?

stefan_2• 7 years ago

Why do you think escalating emissions standards exist? The entire purpose is to gradually retire whole technologies!

It is essentially deprecation. The ultimate goal is always reduce to zero, you will find very few emissions in our continuous world that are somehow safe at arbitrary cutoff X but unsafe above.

pkaye• 7 years ago

So did the recall all those failing 97% of 250 car models in the EU?

armada651• 7 years ago

Yes, but in many countries like the UK the recall is voluntary, it's just a recommendation.

thepangolino• 7 years ago

Thank God it is as it worsens mileage.

walrus01• 7 years ago

I've seen claims that VW's whole diesel emissions and engine control subsystem was actually contracted out to Bosch. But VW execs knew about the software defeat device.

pintxo• 7 years ago

Afaik, yes. bosch was contracted to provide the engine control modules, including features to automatically disable the emission reduction measures. Boach apparently wrote in the contracts that such disablig was only to be used for internal testing, not in production.

_ph_3• 7 years ago

The controlling unit indeed comes from Bosch, but it is a generic unit that can control any kind of engine. The car manufacturer using this unit programs it to perform the actual world, mostly by specifying parameter sets for its operation. The car manufacturer has for example to determine, how exactly the ignition needs to be set up for a given temperature and load etc. There are plenty sensor inputs these profiles can depend on like engine temperature, revs, load, outside temperature, speed of the car. The Bosch software can offer the "test dectection" as one of the inputs. There can be quite reasonable use cases for that information, but that VW used it in the way they did, was VWs choices. The same or a similar control from Bosch unit is probably used by many other car manufacturers, but without abusing this specific "feature".

As the article writes, there are other "strange" engine configurations with other car manufacturers, like using the outside temperature. Reducing or switching off the exhaust cleaning system at outside temperatures below 17C (the tests are guaranteed to be done at 20C) achieves a similar effect to what VW was doing without literally breaking the law by using a defeat device.

pedrocr2• 7 years ago

There can be quite reasonable use cases for that information

What reasonable use case is there for a detection of the test cycle just from ECU information? If you're just testing stuff surely you have access to the ECU to just set a flag manually for anything you need?

yeukhon• 7 years ago

Netflix'a Dirty Money documentary Ep 1 covers this topic: https://www.netflix.com/title/80118100

Yaggo2• 7 years ago

European here. My next car will be electric, likely a Tesla. The dieselgate is the reason why I'm not even considering Volkswagen's upcoming 'ID' family of electric cars. I've similar feeling about the whole traditional auto industry.

adrianN4• 7 years ago

Unfortunately you seem to be in the minority. People are still happily buying VW.

semi-extrinsic• 7 years ago

Why "unfortunately"? Would it be better if every company that does a bad thing was forced to just close shop?

If that be the case I'm pretty sure we'll have to shut down Tesla within a couple of years over their "Autopilot"/"Full Self Driving" claims.

3o4xkp• 7 years ago

I don't care about whether things are "ecologically sound" or not.

I know the earth is important and blah blah, but I have bigger fish to fry every day. I'm sure most people feel like that.

justasimpleman4• 7 years ago

My next car will be "no car at all". Even electric cars have huge externalities that you do not pay for, but your children, namely, the disposal of the batteries, pollution created by the manufacturing, costly extraction of rare earths etc. Instead we should heavily invest in high-speed trains.

lb1lf2• 7 years ago

High-speed trains are wonderful where the population density is sufficiently high, but in more sparsely populated parts, the car (and, for longer distances, the plane) is likely the only viable solution for decades to come.

With that as a starting point, the cars and planes might as well be electric. (Or hydrogen powered!)

maccam94• 7 years ago

Except lithium ion batteries are very recyclable, and rare earth elements aren't in limited supply, they're just usually very sparsely distributed.

xedeon7• 7 years ago

Instead we should heavily invest in high-speed trains.

This is ideal for obvious reasons. If only cities and governments (especially in the U.S.) would see the value. But alas it seems like only some countries can pull it off.

Also, most EV batteries today are highly recyclable and with fewer moving parts they can last much longer than ICE cars.

Dumblydorr• 7 years ago

I love investment in trains but in the USA, small improvements to investment wouldn't be close to enough. We need EVs because the political will is simply not there for any infrastructure spending. Its either EVs or more emissions, sad state of affairs.

jordanbeiber2• 7 years ago

This is ideal.

Unfortunately the society has been pushed into “car as a requirement” territory for ages.

I commend the effort, and I’d like to not need a car... perhaps I don’t, but it will require a few major changes.

It’s crazy that car pools are not more common, btw.

_ph_3• 7 years ago

My next car will be electric too, but I won't exclude VW entirely. They did the most obvious cheating, but basically all Diesel cars would use some schemes to pass the exhaust tests. Their ID line seems to be a genuine attempt at going full electric. I would of course triple-check any claims about their cars going forward.

jordanbeiber2• 7 years ago

I drove a VW passat (bio/ngt gas hybrid) for years.

This stuff pissed me of so much that together with the Exxon stuff I will never own a VW or a fossil ICE car ever again.

I went for the Leaf as I can fit a double buggy in the trunk and it has a good enough range. Not cheap but it’s what I can do as a consumer in this liberal market economy.

Good job, greedy bastards! :)

Side note - using an EV have given me for the third time kind of an technology epiphany experience:

1. Internet

2. The first iphone

3. Now, the EV

Charging in the garage over night, never visting a gas station other than for coffe on the road. It’s awesome!

What’s so incredibly convenient is that the grid is built out already, i.e the infrastructure is in place, and has been for decades.

Imagine someone suggesting we haul this toxic and highly flammable stuff called “gasoline” around the world to “stations”... what a ludicrous idea!

I see parallells to music and media distribution and the internet.

zafka• 7 years ago

This whole imbroglio was a painful learning experience for me. For several years before I bought my used jetta TDI, I started looking into Bio Diesel. I still have 12 Jatropha Curcas trees in my back yard, which were to be the start of a possible startup. I am greatful for the learning experience, but I keep waiting for the wisdom that is supposed to make up a little for the aggravation of aging.

fulafel12• 7 years ago

Teslas are still huge in mass (2000+ kg) and space/dimensions compared to what's needed from a car, and have a correspondingly big carbon footprint from building the thing and batteries. We should aim for 500 kg cars. We could do a 4 person 500 kg car in 1957 after all (Fiat 500).

Someone1234• 7 years ago

We should aim for 500 kg cars.

So you want a Tesla without batteries?

A single battery cell weights approx. 200 lbs (90 kg). The Model 3 has four. The total weight of the batteries is thought to exceed 1000 lbs (inc. housing, harnesses) or about 500 kg.

Then add on all the legally mandated safety technology a vehicle from 1957 wouldn't have (front airbags, seat belts, windshield defroster/washer, taller seats, and much higher crash absorption/roll support).

You could absolutely build a 500 kg kit-car that has 1957 safety. But no commercial vendor can sell you a vehicle even in that ballpark.

Yaggo2• 7 years ago

Model 3 weights 1600...1800 kg (SR/LR), not much more than similar spec'd ICE car.

carboy• 7 years ago

I could never understand how something that was less refined, diesel, meaning that it had more unnecessary “stuff” could produce less emissions. Diesel always did, and always will, as far as I’m concerned he snake oil when it comes to emissions.

benj111• 7 years ago

Its less refined in the sense of less cracking.

It contains more energy for a given volume, so you get better fuel economy than with petrol. It might not be better for the particulates and NOx that we are now optimising for, but its better for CO2, that we were optimising for. So I don't think the snake oil label is warranted.

lscotte2• 7 years ago

Diesel engines are also significantly more efficient at converting that energy into usable torque. That's why an idling diesel consumes almost no fuel - it isn't doing any work.

CzechTech• 7 years ago

I always cringe when one of these "we could have been driving electric cars" articles shows up. Cars are not bread and the vast vast majority of the planet's population (who even own a car) don't just go around shopping for new cars. It's going to take DECADES before electric cars have a significant impact on the climate change. That is an economic reality. (Not to mention that electric cars are only now starting to barely match gasoline cars in terms of usability and range.)

jokoon3• 7 years ago

I live in france, and I often spot cars that have a grey/blue smokes. Many other cars also have this awful smell.

Here, every 2 years, cars must legally be examined in a "controle technique", like an exam which controls several aspects of a vehicle.

I wonder if those fumes are legal, and I really wonder if people can manage to cheat at that exam, because I think I'm noticing an epidemic of those fumes.

How can I really be expected to ride a bike next to cars like this? I'd rather spend time in a tram or bus than expose my lungs to this.

Al-Khwarizmi2• 7 years ago

In Spain we have a similar exam and unfortunately I know that there are indeed people that cheat. I don't know the details but there are ways to make their car emit less smoke only that day, and then go back to "normal".

Probably we should do as many Asians do in this, and make it normal to go out with a face mask.

Not sure what kind of masks have to be used to effectively not breathe pollution, and their effectivity, though. Getting informed about that is in my to-do list.

jokoon3• 7 years ago

no mask can filter particles

eebynight• 7 years ago

I have an older diesel VW. It is extremely common for people to run tunes that disable the check engine light that is produced when you delete the exhaust gas recirculation valve (EGR), urea pump and diesel particulate filter (DPF).

For things like the EGR, blocking it off is simply installing a bracket or two and closing off some vacuum lines. I can imagine people would simply reinstall this and swap the tune to pass and then immediately remove it after.

In the US, for emissions, all you have to do is not run any check engine lights when at the test center which is easy to do in the software tune.

UncleEntity2• 7 years ago

In the US, for emissions, all you have to do is not run any check engine lights when at the test center which is easy to do in the software tune.

Which is very disappointing...

When I first moved to Arizona I had a '67 Galaxy and that thing got to run on the dyno in the emissions office but when I got my truck to replace it (not having AC in Arizona isn't fun) all they did was plug in into the computer and charge me more for the test.

My (carburated) motorcycle was in between, they couldn't run it on the dyno and couldn't plug a computer into it so just tested it at idle -- which seemed pretty pointless to me, even being all hopped-up it still got better gas mileage than most cars on the road, though it did get pretty horrible gas mileage for a motorcycle considering I had to run premium in it.

sidcool• 7 years ago

This happens in almost all industries. Big oils' repetitive attempts to kill EVs, Big Pharmas, insurance and hospitals lobbying against healthcare reforms, Big Banks working against finance regulations etc.

hjk05• 7 years ago

Big Parma is lobbying FOR healthcare reforms.

maxxxxx• 7 years ago

Not for anything that may reduce profits though.

How the car industry hid the truth about diesel emissions

Related Stories