VW cheat emissions test with "defeat device"

[bhall II]

My feelings on this topic are summarized quite well by the following post:

[But in all of this the most common statement is that " VW car swere shown to produce 40 times, or up to 40 times the permitted EPA Nox level.In the entire WVU report I cannot find one reference to 40 times the EPA Nox limit.]

Those references - "...emissions of NOx increased by a factor of 10 to 40 times above the EPA compliant levels..." - come from EPA's notice of violation, not from the CAFEE (WVU) report.

And I meant to ask this earlier...

The same cannot be said of VW's bypass algorithm, because its functionality is completely irrelevant. The mere fact that it exists undeclared is the violation, and that would remain true even if it somehow improved emissions in everyday driving.

I am tempted to reply that the law is an ass, but I guess most people know that already.

...why is the law an ass? It seems to me it was written with as little ambiguity as possible.

You've seen the debate here about how emissions standards are ultimately defined, whether it's by rule or by the rule's testing procedures. As far as enforcement and penalties are concerned, the Clean Air Act sidesteps that debate entirely.

Volkswagen violated the CAA due to a failure to disclose all "elements of design" that control vehicle emissions, which is a binary requirement: either it's there, or it's not. Everything else is just evidence of the violation.

[Tommy Cookers]

imo this is an informative gem
http://www.greencarcongress.com/2015/09 ... -vw2l.html


btw a post to joe saward's site shows that meat-agriculture is the great producer of NOx

regarding the illegality of a 'defeat device' (ie anything whose main or only purpose is to degrade emissions in off-test use) ...
a device or design feature eg VVT or other feature is not a DD as legally defined but will substantially degrade emissions off-test

[DaveW]

My feelings on this topic are summarized quite well by the following post:

[But in all of this the most common statement is that " VW car swere shown to produce 40 times, or up to 40 times the permitted EPA Nox level.In the entire WVU report I cannot find one reference to 40 times the EPA Nox limit.]

Those references - "...emissions of NOx increased by a factor of 10 to 40 times above the EPA compliant levels..." - come from EPA's notice of violation, not from the CAFEE (WVU) report.

The NOV is an accusation, not a statement of fact (though it become factual if/when more information surfaces).

And I meant to ask this earlier... why is the law an ass? It seems to me it was written with as little ambiguity as possible.

That is based on the thought that most control software comprises a series of algorithms connected by conditional statements (e.g. if .. then begin .. end else begin .. end;). Indeed algorithms may also contain conditional statements. It could (would) be argued that each and every conditional statement is a potential "cheat device".

The consequence of that is that all control software must be published with full explanations. Power train designers would love that as an idea, and the law would not know what to do with the consequences.

[bhall II]

The consequence of that is that all control software must be published with full explanations. Power train designers would love that as an idea, and the law would not know what to do with the consequences.

The change of tense in your phrasing here is a bit confusing, because I can't tell if you've offered commentary on the existing rules or speculation about what you think those rules should be.

But, that is, in fact, the current rule, and it's how VW got caught. In real-world conditions, their emissions systems didn't perform as described within the relevant application(s) for certification.

[Edis]

imo this is an informative gem
http://www.greencarcongress.com/2015/09 ... -vw2l.html


btw a post to joe saward's site shows that meat-agriculture is the great producer of NOx

regarding the illegality of a 'defeat device' (ie anything whose main or only purpose is to degrade emissions in off-test use) ...
a device or design feature eg VVT or other feature is not a DD as legally defined but will substantially degrade emissions off-test

Agriculture is a minor source of NOx, in Europe transport is responsible for the largest share of NOx emissions by 46% of total emissions. This is followed by the energy sector and then industry. However, NOx emissions from transport is declining faster than NOx emissions from other sources.

[bdr529]

imo this is an informative gem
http://www.greencarcongress.com/2015/09 ... -vw2l.html

btw a post to joe saward's site shows that meat-agriculture is the great producer of NOx

Volkswagen's got that covered, I think their called Volks Wurst or Wurstwagen, they also make ketchup
Personally I prefer the "Smoky bacon" but the "Curry bockwurst" does sounds delicious

[DaveW]

The change of tense in your phrasing here is a bit confusing, because I can't tell if you've offered commentary on the existing rules or speculation about what you think those rules should be.

Apologies for my grammar. Neither of your options is the case. What I tried to say is if an (unspecified) "defeat device" must always be declared, then the whole code must be made public, with documentation, to be sure that any "defeat device" is covered. Anything less would be a lawyer's dream.

But, that is, in fact, the current rule, and it's how VW got caught. In real-world conditions, their emissions systems didn't perform as described within the relevant application(s) for certification.

The only documents I have read so far a) published the evidence, and b) issued a "Notification of Violation". Your "got caught" statement may well be true, but it is a step on from that evidence, I think. "Guilty until proven innocent", perhaps.

[Tim.Wright]

The reason the EPA notice was served was that after a lot of testing and dialogue, VW eventually threw their hands up and admitted the defeat device. So there was already an admission of guilt before the notice was served.

[DaveW]

Quote from http://money.cnn.com/2015/09/20/autos/v ... o-apology/

Volkswagen's CEO apologized for customers' "broken trust" on Sunday after the company was accused last week of defrauding environmental regulators.
Martin Winterkorn issued a statement that stopped short of admitting guilt, but said he was "deeply sorry that we have broken the trust of our customers and the public."

[Tim.Wright]

That was a public relations announcement - so it was softened.

Their admission of guilt is documented in the EPA's notice of violation:

It became clear that CARB and the EPA would not approve certificates of conformity for VW's 2016 model year diesel vehicles until VW could adequately explain the anomalous emissions and ensure the agencies that the 2016 model year vehicle's would not have similar issues. Only then did VW admit that it had designed and installed a defeat device in these vehicles in the form of a sophisticated software algorithm that detected when a vehicle was undergoing testing.

The epa doc explains that the whole saga started many months ago due the results of an unrelated study. When pushed for an answer VW responded with a lot of handwaving and then issued a new version of the engine software. This was tested again and found to be still insatisfactory at which point they admitted the defeat device.

[djos]

Quote from http://money.cnn.com/2015/09/20/autos/v ... o-apology/

Volkswagen's CEO apologized for customers' "broken trust" on Sunday after the company was accused last week of defrauding environmental regulators.
Martin Winterkorn issued a statement that stopped short of admitting guilt, but said he was "deeply sorry that we have broken the trust of our customers and the public."

You misunderstand, that article is talking about personal guilt, he's adamant he knew nothing about it.

[bhall II]

What I tried to say is if an (unspecified) "defeat device" must always be declared, then the whole code must be made public, with documentation, to be sure that any "defeat device" is covered. Anything less would be a lawyer's dream.

That is, indeed, one element of current requirements...

(a) All information listed in this section must be submitted to the Agency according to the requirements specified in § 86.1843-01.

[...]

(11) A list of all auxiliary emission control devices (AECD) installed on any applicable vehicles, including a justification for each AECD, the parameters they sense and control, a detailed justification of each AECD which results in a reduction in effectiveness of the emission control system, and rationale for why the AECD is not a defeat device as defined under §§ 86.1809-01 and 86.1809-10. For any AECD uniquely used at high altitudes, EPA may request engineering emission data to quantify any emission impact and validity of the AECD. For any AECD uniquely used on multi-fuel vehicles when operated on fuels other than gasoline, EPA may request engineering emission data to quantify any emission impact and validity of the AECD.

Auxiliary Emission Control Device (AECD) means any element of design which senses temperature, vehicle speed, engine RPM, transmission gear, manifold vacuum, or any other parameter for the purpose of activating, modulating, delaying, or deactivating the operation of any part of the emission control system.

The existence of AECDs not specified in VW's applications for "certificates of conformity" is ultimately the basis for EPA's case against VW. As you said, other requirements open the door to miles of speculation. Here, matters are greatly simplified: either it's there, or it's not. (The EPA is criminally underfunded by Congress and often grossly mismanaged by administrators, which has resulted in rules that rely on a great deal of self-policing and spot checks of that self-policing.)



Everything else is evidence of the violation, and that's the part the media seems to be getting wrong or leaving out. I guess talk of auxiliary emission control devices isn't as sexy as "...emissions of NOx increased by a factor of 10 to 40 times above the EPA compliant levels..."

[Ultra]

Eric Peters sums it quite nicely:

Print Friendly
This could kill VW – until recently (until last week) the world’s largest car company.VW lead 1

But unlike say the exploding Pinto fiasco this is not a story about defective cars. It is a story about defective public policy.

None of the VW cars now in the crosshairs are unreliable, dangerous or shoddily built. They were simply programmed to give their owners best-case fuel economy and performance. Software embedded within each vehicle’s computer – which monitors and controls the operation of the engine – would furtively adjust those parameters slightly to sneak by emissions tests when the vehicle was plugged in for testing. But once out on the road, the calibrations would revert to optimal – for mileage and performance.

Now, the hysterical media accounts of the above make it seem that the alteration via code of the vehicles’ exhaust emissions was anything but slight. Shrill cries of up to “40 times” the “allowable maximum” echo across the land.

Well, true.

But, misleading.

Because not defined – put in context.VW 2

What is the “allowable maximum”?

It is a very small number.

Less than 1 percent of the total volume of the car’s exhaust. We are talking fractions of percentages here. Which is why talk of “40 percent” is so misleading and, frankly, deliberately dishonest.

Left out of context, the figure sounds alarming. As in 40 percent of 100 percent.

As opposed to 40 percent of the remaining unscrubbed 1-3 percent or .05 percent or whatever it is (depending on the specific “harmful” byproduct being belabored).

The truth – explained rarely, for reasons that will become obvious – is that the emissions of new cars (and recent-vintage cars) have been so thoroughly cleaned up they hardly exist at all. Catalytic converters (and especially “three way” catalytic converters with oxygen sensors) and fuel injection alone eliminated about two-thirds of the objectionable effluvia from the exhaust stream – and they’ve been around since the 1980s. Most of the remaining third was dealt with during the ’90s, via more precise forms of fuel delivery (port fuel injection replaced throttle body fuel injection) and more sophisticated engine computers capable of real-time monitoring and adjustment of parameters, and of alerting the vehicle’s owner to the need for a check (OBD II).

Since the late ’90s/early 2000s, the industry has been chasing diminishing returns. The remaining 3 percent or so of the exhaust stream that’s not been “controlled.”VW 3

You may begin to see the problem here.

Internal combustion is always going to produce some emissions. The engineers have picked the low hanging (and mid-hanging) fruit. But the EPA insists on what amounts to a zero emissions internal combustion engine.

Which, of course, is impossible.

Which may be just the point.

Set unattainable standards – then denounce the victim for “noncompliance.”

VW’s sin was trying to get diesels that people would want to buy into the showrooms. These would be diesels that went farther than an otherwise-equivalent gas-engined car on a gallon of fuel to offset the higher up-front cost of buying the diesel-powered vehicle. Or at least, far enough – relative to the gas-engined equivalent – to justify the price premium.

People also expected – demanded – that the vehicles perform. That they accelerate when the accelerator is pushed.atlas shrugged image

VW set the calibrations to deliver those things. The operating characteristics its customers want.

VW is in hot water because of that. Because it put customers – rather than government – first.

No one has alleged that any of the “affected” vehicles runs poorly. The fact is they run better than they would have if VW had set the calibrations to appease the implacable EPA.

Which will never be appeased until we’re all driving $60,000 “zero emissions” electric cars we can’t afford. Which will put most of us into public (that is, government) transport. If we’re transported at all. Probably, we’ll be herded into urban cores, stacked like proles – for the sake of “the environment.”

It is a tragedy of stupidity and maliciousness and engineering ignorance.

Consider, for instance, the fact that if it were not for federal “safety” mandates, VW (and other car companies) would be able to sell vehicles hundreds of pounds lighter than the current average. Which, in turn, would allow for smaller engines – which burn less fuel. Which, in turn produce a lesser volume of exhaust. Even if a hypothetical 1,600 pound ultra-light vehicle’s exhaust stream were, let’s say, 2 percent “dirtier” than a current 2,300 pound EPA (and DOT) approved “safety” car’s, if the ultra-light burns 40 percent less fuel, its total output is still much lower than then government-approved car’s.atlas shrugged 2

But such cars (the ultra-lights) have – effectively – been legislated out of existence.

At the same time, the cars that may still be manufactured are required to meet increasingly unattainable standards, putting the manufacturers (like VW) in the position of manufacturing government-compliant cars that cost too much and perform poorly that few will want to buy… or “cheating” the government, in order to build cars people will actually want to buy.

What’s happening to VW could have come right out of Atlas Shrugged, Ayn Rand’s cumbersome but nonetheless predictive novel of 50 years ago. VW cast as the real-life version of Rearden Steel.

Some inside baseball: Mazda has been trying to get its Sky-D diesel engine EPA-compliant (while also customer-viable) for the past two years, without success so far. You are denied this 50-plus MPG (and extremely clean) diesel because of the particulate jihadists in Washington.

Remember: In neither case (VW or Mazda) are we talking about a return to the LA of the early ’70s, a feasting on lead chip paints and bathing in DDT. It’s all a bogey at this point. A straw man. A phantom, meant to scare you. But it has no reality.

The “emissions problem” has been solved – decades ago. But the EPA, et al, cannot admit this.

Because then there’d be no need for the EPA.

http://ericpetersautos.com/2015/09/23/c ... -by-uncle/

[bhall II]

Eric Peters sums it quite nicely:

[100% pure, unadulterated, Grade A bullshit]

http://ericpetersautos.com/2015/09/23/c ... -by-uncle/

For as much as he hammers the media about its misrepresentation of the facts, his rampant obfuscation is laughable. The issue here is not now, nor has it ever been, about the increased levels of NOx emissions from VW diesels in everyday use; it's about the company's widespread deception of consumers and government regulators.

The company was offered a myriad opportunities to correct the problem. It did not.

The company was warned against using the "defeat device" by its supplier in 2007. It did anyway.

The company was warned against cheating emissions standards by one of its own engineers in 2011. It did anyway.

(And this case doesn't mark the first time VW has had a run-in with EPA over "defeat devices.")

If the free market didn't incentivize corporate malfeasance, then there would be no need for regulatory bodies to protect the public from acts of corporate malfeasance. Instead, Volkswagen's actions have yet again demonstrated the need for oversight.

Despite raking in nearly €200 billion in revenues for 2014, inlcluding over €11 billion in profits, VW's blind ambition still made it necessary to cut corners. That's symptomatic of the ever-widening gulf that currently exists between responsibility and profitability.

[turbof1]

If the free market didn't incentivize corporate malfeasance, then there would be no need for regulatory bodies to protect the public from acts of corporate malfeasance. Instead, Volkswagen's actions have yet again demonstrated the need for oversight.

As an economist by academic education, I very much salute you for that comment.