Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Government Software

EPA To Overhaul Emissions Testing In the Wake of VW Cheating 203

New submitter kheldan writes with this snippet from The Consumerist: A week after ordering Volkswagen to recall 500,000 vehicles that contain "defeat devices" designed to cheat emissions tests, the Environmental Protection Agency announced it would overhaul its compliance processes to ensure vehicles meet standards not only in controlled environments but in real-world driving conditions, and adds What may be the story-behind-the-story here, are the two Elephants in the Room: One, how many other automakers in the world have been 'gaming' the system like German automakers apparently have been all along, and Two, are these changes to the certification process at the USEPA going to 'trickle down' to the state and local levels, affecting routine emissions testing of individual vehicles? Questions peripheral to these may include: How much is this going to affect new vehicle prices in the future, and how much is this going to affect the fair market value of used vehicles?
This discussion has been archived. No new comments can be posted.

EPA To Overhaul Emissions Testing In the Wake of VW Cheating

Comments Filter:
  • by goodmanj ( 234846 ) on Saturday September 26, 2015 @05:47PM (#50604951)

    First let me say that this change is urgently needed.

    But, it's unlikely that automakers who build gasoline cars are cheating like VW did. It's especially difficult to clean NOx from diesel engine exhaust because unlike gasoline engines, the exhaust contains lots of extra oxygen. Diesels need special NOx-cleaning devices which add cost and weight, and can seriously limit performance in some situations. Gasoline engines just need minor modifications to the engine computer software and the catalytic converter to clean NOx, so there's very little need to cheat.

    https://en.wikipedia.org/wiki/... [wikipedia.org]
    https://en.wikipedia.org/wiki/... [wikipedia.org]
    http://www.google.com/patents/... [google.com]

    • by drnb ( 2434720 ) on Saturday September 26, 2015 @06:03PM (#50604979)

      It's especially difficult to clean NOx from diesel engine exhaust because unlike gasoline engines, the exhaust contains lots of extra oxygen. Diesels need special NOx-cleaning devices which add cost and weight, and can seriously limit performance in some situations.

      No. It is no longer especially difficult because VW and other diesel makers have already solved this problem. Every cheating VW diesel on the road can have low polluting exhaust. The hardware and software is there, already installed and operating. That is how they pass emissions tests, the software enables all the emissions controls. How the software cheats is to turn off the emission controls if it looks like someone is actually driving.

      The fix is a simple software patch to stop turning off the emissions controls.

      The downside is that performance and mileage will be reduced.

      • by Nutria ( 679911 )

        How the software cheats is to turn off the emission controls if it looks like someone is actually driving.

        Does that mean "wheels spinning" as opposed to just "engine revving"?

        • by drnb ( 2434720 )

          How the software cheats is to turn off the emission controls if it looks like someone is actually driving.

          Does that mean "wheels spinning" as opposed to just "engine revving"?

          Not necessarily. It could mean looking for driver inputs. Steering wheel, accelerator pedal, brake pedal, etc. Things characteristic of actually being on the road.

        • by goodmanj ( 234846 ) on Saturday September 26, 2015 @06:36PM (#50605059)

          It's very clever (but evil): EPA says the software looks at a variety of factors, including wheel speed, steering wheel position, engine run time, and barometric pressure (!), and compares those data against EPA's published testing guidelines.

          • by Nutria ( 679911 )

            including wheel speed, ... and compares those data against EPA's published testing guidelines.

            Wow. That's not some "minor misfeature"....

            • by goodmanj ( 234846 ) on Saturday September 26, 2015 @07:30PM (#50605217)

              No, even when breaking the law, German engineering does not screw around.

              • by Nutria ( 679911 )

                even when breaking the law, German engineering does not screw around.

                Good thing! Made it easy to catch them in 1945, and will also do so 70 years later...

            • Of course not. It had to be quite advanced to detect the difference between car on a dyno in an emissions test chamber and car on a dyno being performance tested.

          • It's very clever (but evil): EPA says the software looks at a variety of factors, including wheel speed, steering wheel position, engine run time, and barometric pressure (!), and compares those data against EPA's published testing guidelines.

            This sounds like the required Federal testing for students. The Feds give out the information that children must know, and so the administration teaches the children that information.

            • This sounds like the required Federal testing for students. The Feds give out the information that children must know, and so the administration teaches the children that information.

              Are you familiar with the FAA written exams for private pilots? Passing score is 70% and 80% of the questions are known ahead of time. These questions are from exams from previous years, unchanged, and these exams are available for study and practice. The 20% of questions that are "new", largely old questions with the given numbers changed.

              • This sounds like the required Federal testing for students. The Feds give out the information that children must know, and so the administration teaches the children that information.

                Are you familiar with the FAA written exams for private pilots? Passing score is 70% and 80% of the questions are known ahead of time. These questions are from exams from previous years, unchanged, and these exams are available for study and practice. The 20% of questions that are "new", largely old questions with the given numbers changed.

                Yes, studied past exams and took the test and got a perfect score on the private pilot and then later on the instrument rating. Passed the checkride for Private Pilot first time. Took two tries on the Instrument rating. Turns out studying to the test doesn't correspond to doing well in real world scenarios.

          • by rtb61 ( 674572 )

            The easiest method for everyone, electric cars and the EPA can do it's regulating thing at the power station. You know it is coming, want to be a pollution pus hog, then you will be paying a large regulatory cost to do so.

            • The easiest method for everyone, electric cars and the EPA can do it's regulating thing at the power station. You know it is coming, want to be a pollution pus hog, then you will be paying a large regulatory cost to do so.

              Yeah but... VW wanted to beat Toyota out worldwide. hybrids and electrics have become Toyota's playground, so VW decided to stay with the German companies' playground, diesels. People have been noticing for years that diesels are giving mileage results comparable to hybrids. But VW wanted their diesels to offer no user abrasion greater than a gasoline car; and filling up a tank of urea now and then seemed like something undesirable.

        • How the software cheats is to turn off the emission controls if it looks like someone is actually driving.

          Does that mean "wheels spinning" as opposed to just "engine revving"?

          emissions testing involves a chassis dynamometer with rolling drums for the wheels, doesn't it? And maybe variable loads?

      • by goodmanj ( 234846 ) on Saturday September 26, 2015 @06:32PM (#50605041)

        You're right that the technology is installed and works, but the performance and maintenance downsides are severe. I didn't want to make my original post too long to read, but here's more detail to explain.

        Most small VWs (Jetta, etc) use a "lean NOx trap" to capture NOx in a zeolite sponge. The zeolite fills up with NOx and needs to be cleaned out periodically (every minute or two, takes a few seconds). During the cleaning cycle, engine power is limited to *20%* of maximum. VW's patent says they wait until the driver eases off on the throttle to do it, but still, that's a huge performance hit and a big incentive to cheat (by not doing the cleaning.)

        See patent link in my original post for details (warning: machine-translated from German.)

        In VW's larger vehicles (Passat, mostly), the car carries an extra tank full of gallons of urea, which is sprayed into the exhaust to react with the NOx. This reaction needs precise temperature controls (which probably limits engine performance), and the tank is big and heavy. By using less urea than needed, VW can use a smaller lighter tank, which needs to be refilled less often. (VW pays for urea replacement for the first 30,000 miles.)

        • by Sorny ( 521429 ) on Saturday September 26, 2015 @07:58PM (#50605305) Homepage

          VW's patented setup sounds far less ideal than the well understood SCR/DEF setup everyone else uses. 20% of max power to burn out the NOx trap? No way in hell would I want that!

          The precise temperature controls you allude to for SCR/DEF is hogwash. You've got quite a bit of leeway to get up to temp before the system starts dosing before throwing codes and going into limp mode (think like over an hour of operation). If you didn't, then all of us in places with 4 real seasons would have diesels that wouldn't go anywhere because DEF freezes at about 18 degrees F. Where I live, we can go months with temps lower than that, and it takes time to thaw the DEF so it'll flow.

          DEF consumption is 1-3% of fuel consumption. So figure 1-3 gallons of DEF used for every 100 gallons of diesel; that's for the puny little 2 liters in a VW all the way up to the 13L monsters in a Peterbilt... 1-3% of fuel, like clockwork. How many gallons of DEF would be needed to go the oil change interval on a VW TDi? Not many.

          Lastly, on DEF, any fool paying stealership prices deserves to get ripped off. Drive to a truck stop and enjoy ISO rated DEF/AdBlue at $2.70 or so a gallon. My last DEF fill cost me a whopping $16 for my truck at just over 6 gallons in ~8k miles.. A 3.0L diesel truck that gets 24-27 MPG in mixed driving. A 2.0L TDi should be able to go 12k+ miles on 5 gallons considering how much less fuel they burn than my truck does.

          • DEF consumption is 1-3% of fuel consumption

            Thanks, I was looking for this figure for an article I'm writing on diesel NOx chemistry. Sounds like you've got first-hand experience, but do you have an online source I could cite for that? A service manual or something?

          • by tlhIngan ( 30335 )

            VW's patented setup sounds far less ideal than the well understood SCR/DEF setup everyone else uses. 20% of max power to burn out the NOx trap? No way in hell would I want that!

            Most car engines run around that power - it's why car engines are small and powerful compared to many other engines. Think aviation and truck engines - sure those engines are big, but they also run at higher power settings for longer periods of time. Aviation engines especially - they're extremely big for the power. (160 cu. in., or

            • by Mal-2 ( 675116 )

              Aviation engines especially - they're extremely big for the power. (160 cu. in., or over 5L, and it produces a mere... 140hp?).

              Check your conversions again. 5 liters is about 300 cubic inches, as anyone with an old-school V-8 is likely to know. Thus, one or the other of your numbers is off. 160 cubic inches is not a large engine.

            • Anyhow, I think the problem was VW didn't want to use an AdBlue style system.

              But they *do* also use an AdBlue system on their larger vehicles (notably, the Passat), and the EPA claims they were cheating on the Passat too. Quite frankly, I have no idea why.

        • That's a good case for diesel-electric hybrids. During the cleaning cycle, supplement the 20% engine power with electric locomotion.
        • by caseih ( 160668 ) on Saturday September 26, 2015 @10:25PM (#50605683)

          Actually, no. Urea does require hotter temperatures, true, but it surely doesn't impact performance. The way diesel pollution works is that you can either lower compression and combustion through EGR to reduce NOx, but this tends to produce particulates and reduces fuel efficiency. Or you can increase efficiency and run the engine hotter, possibly with more compression, which virtually eliminates particulates, but hotter combustion temperatures increase NOx production.

          If anything Urea lets the engine run a lot closer to its more efficient state with more compression and higher temperatures. As you say the urea plus the catalyzing exhaust chamber does add weight. But the biggest problem is the availability of urea (in north America) and the handling of it. Especially in the winter.

          We run a machine on the farm with Tier 4I emissions on it, and every year we buy about 800 L of urea. It's about $1 CAD/L. So it does add overall cost, though to put it in perspective, it costs nearly $400 a day in diesel fuel during harvest for the same machine, totaling $800 a day for the two machines. But this engine is also more efficient than previous models, so fuel consumption is lower. We don't run the machine in the winter so we've never had any problems with it gelling, and we've never had the machine derate due to urea problems. In my mind, urea injection is really the only practical way to produce cleaner diesels. This is important with biologically-derived fuels as well, such as biodiesel. The carbon cost of urea production and handling probably makes it a wash in terms of CO2 emissions, despite higher efficiency engines. Urea is made from natural gas reformation.

          • Thanks for the info. I'm coming at this from basic chemistry and engineering knowledge and by reading up on the technology and VW's patents, but I don't have first-hand experience with these engines. The big question for me is, "what's the upside for VW?" For the lean NOx trap system, there's an obvious performance hit, but the EPA says VW cheated on its urea-system Passat too, and I can't figure out why. I had guessed the temperature restrictions might be part of it, but if that's not the case I'm left

            • I think the urea cheating may be for reasons of customer acceptability.

              Commercial diesel equipment which uses urea as NOx control can use between 1-5% of the volume of diesel fuel in urea solution. In Europe, where urea has been used in trucks for many years, a truck stop will have urea pumps next to the diesel pumps. So that you can fill both tanks simultaneously.

              For passenger cars, there is considerable pressure to make the urea a service item, which can be refilled at the 10k mile service interval,
              • Yeah, that's been my best guess too. But (10,000 miles / (40 miles/gallon diesel)) * 2% urea/ diesel = 5 gallons urea per service interval. Passat has a 5-gallon [cars.com] urea tank, which is just the right size.

                So they're not skimping on size/weight of the tank. The only explanations left that I can see: either the urea system puts some performance limits on the engine that I haven't figured out yet, or else they just want to save some money on urea (since they pay for it for the first 30,000 miles.) But risking

                • by caseih ( 160668 )

                  There really is resistance to urea here in NA. It's seen as a burden by just about everyone and it really does play into consumers' buying decisions. $20 worth of chemical can translate into thousands of dollars in lost sales. Especially in a market dominated by gasoline cars.

                  It's also entirely possible that in real-world conditions the EPAs regulations are simply unattainable in any acceptable way. Now that the EPA is going to have to move to real-world testing, this could be a good thing to let the gov

              • If the "1-3% of fuel use" metric is accurate, then a 10,000 mile service interval on a 40 MPG TDI would require a tank that could hold 2.5-7.5 gallons of urea. That's not too bad.

                (Of course, I'd rather have a 5,000 mile service interval -- which you'd need anyway if you run biodiesel, since more of it makes it past the piston rings and pollutes the oil -- and a urea tank half the size.)

              • I think the urea cheating may be for reasons of customer acceptability. Commercial diesel equipment which uses urea as NOx control can use between 1-5% of the volume of diesel fuel in urea solution. In Europe, where urea has been used in trucks for many years, a truck stop will have urea pumps next to the diesel pumps. So that you can fill both tanks simultaneously. For passenger cars, there is considerable pressure to make the urea a service item, which can be refilled at the 10k mile service interval, in order to avoid the risk of customer non-acceptance. There is therefore a strong pressure to be as frugal as possible with the urea in order to minimize the size/weight of the urea tank and ensure that it will last the duration of the service interval.

                Coming soon: do it yourself kits to extract urea from pee, to go with the biodiesel manufacture kits.

        • You're right that the technology is installed and works, but the performance and maintenance downsides are severe. I didn't want to make my original post too long to read, but here's more detail to explain.

          Most small VWs (Jetta, etc) use a "lean NOx trap" to capture NOx in a zeolite sponge. The zeolite fills up with NOx and needs to be cleaned out periodically (every minute or two, takes a few seconds). During the cleaning cycle, engine power is limited to *20%* of maximum. VW's patent says they wait until the driver eases off on the throttle to do it, but still, that's a huge performance hit and a big incentive to cheat (by not doing the cleaning.)

          See patent link in my original post for details (warning: machine-translated from German.)

          In VW's larger vehicles (Passat, mostly), the car carries an extra tank full of gallons of urea, which is sprayed into the exhaust to react with the NOx. This reaction needs precise temperature controls (which probably limits engine performance), and the tank is big and heavy. By using less urea than needed, VW can use a smaller lighter tank, which needs to be refilled less often. (VW pays for urea replacement for the first 30,000 miles.)

          I have my own idea (patent applied for). The exhaust is captured in a large balloon towed behind the car. At the end of the day, the balloon is connected to a pipe at your house, and the contents and downloaded and treated.

    • FWIW if the emissions controls are controlled by software then disabling them to boost performance may be just as plausible a tactic for gasoline engines as diesel engines. Of course the payback may be smaller and so the risk/reward to an automaker may not tempt them so much. However some hobbyists may disable their emissions even for a minor performance boost when modifying their car.
    • by oic0 ( 1864384 ) on Saturday September 26, 2015 @09:30PM (#50605539)
      Motorcycles cheat like crazy. Almost all of them. They severely lean out portions of the fuel map that get used in the test and leave other portions at optimal or even a little rich to reduce warranty claims. First thing most owners do is buy an exhaust system for the sound, often eliminating the cat, and a tuner to fix the horrible EPA test cheater fueling.
    • NOx it's not an issue in gasoline, but CO and CO2 are. And are probably cheated also, for the monster trucks they sell in America. But as it is local industry, audits are softer. And with low taxes on gas, EPA's stance on reducing car emissions is a practical joke.
      Your government, reflecting your society, doesn't take the environment seriously; but you are starting to taste climate change in the form of drought, we'll see if that changes.

    • by mjwx ( 966435 )

      First let me say that this change is urgently needed.

      But, it's unlikely that automakers who build gasoline cars are cheating like VW did. It's especially difficult to clean NOx from diesel engine exhaust because unlike gasoline engines, the exhaust contains lots of extra oxygen. Diesels need special NOx-cleaning devices which add cost and weight, and can seriously limit performance in some situations. Gasoline engines just need minor modifications to the engine computer software and the catalytic converter to clean NOx, so there's very little need to cheat.

      https://en.wikipedia.org/wiki/... [wikipedia.org]
      https://en.wikipedia.org/wiki/... [wikipedia.org]
      http://www.google.com/patents/... [google.com]

      Many Diseasels have a NOx cleaning device, it's called Urea Injection. Its a system that adds a Diesel Exhaust Fluid (DEF) to the Catalytic Converter that reduces pollutants, more specifically it converts NOx into water (H2O) and Nitrogen (N2).

      DEF is more commonly know by it's commercial name, AdBlue.

      VW is pretty much the only manufacturer not to use urea injection (as it adds cost, complexity and maintenance costs which are considered poison to people tight fisted enough to buy diesel passenger cars)

    • First let me say that this change is urgently needed.

      But, it's unlikely that automakers who build gasoline cars are cheating like VW did. It's especially difficult to clean NOx from diesel engine exhaust because unlike gasoline engines, the exhaust contains lots of extra oxygen. Diesels need special NOx-cleaning devices which add cost and weight, and can seriously limit performance in some situations. Gasoline engines just need minor modifications to the engine computer software and the catalytic converter to clean NOx, so there's very little need to cheat.

      https://en.wikipedia.org/wiki/... [wikipedia.org] https://en.wikipedia.org/wiki/... [wikipedia.org] http://www.google.com/patents/... [google.com]

      i'm glad you brought this up because I was wondering just today...
      is it therefore impossible to use catalytic converters on diesels for the NOx like we do on gasoline cars? Is it because there is more to clean up? Or is it because they don't want to clog up the exhaust?

  • "the Environmental Protection Agency announced it would overhaul its compliance processes"

    Would they want to know, who wrote the software and who authorized the writing of the software?
  • by sshir ( 623215 ) on Saturday September 26, 2015 @08:49PM (#50605427)
    I know, it might be unpopular, but consider this explanation: what if that mode was designed to be turned on when car detects running in a badly ventilated area like indoors or in a tunnel and such? Just to avoid becoming a health hazard. And nobody realized that such mode would interfere with EPA tests. And VW own testers were simply replicating EPA testing rig to insure that ther testing is the same, while having no clue how engine works. While it is still probable, that someone in VW realized that there is a problem, they kept their mouth shut for various reasons. But generally this explanation does not require any wide conspiracy or anything.
    • by haruchai ( 17472 ) on Saturday September 26, 2015 @09:19PM (#50605517)

      Nor does it require an expansive mea culpa by the CEO of the actual company, his resignation, the firing of several senior executives and setting aside $7B to remediate the affected vehicles.

      Perhaps you should get one of those vacant VW posts; sounds like you could have saved them several diesel truckloads of money.

      • Ladies and gentlemen, I'll be brief. The issue here is not whether we broke a few rules, or took a few liberties with our EPA tests - we did.

        But you can't hold a whole auto company responsible for the behavior of a few, sick twisted individuals. For if you do, then shouldn't we blame the whole automotive industry? And if the whole automotive industry is guilty, then isn't this an indictment of our industry in general? I put it to you, Greg - isn't this an indictment of our entire American society? Well,

  • Yes, I understand that VW cheated the regulators in terms of emissions. Fair enough. They cheated. They got caught. They should pay the price.

    What I don't understand is end users being upset. In my life, I've purchased 5 cars to date (all new). Never has the emissions level of the vehicle been a factor, at any level, in my decision over which vehicle to purchase. Horsepower? Yes, Fuel efficiency? Yes. Cost? Yes. Emissions? Not even on the radar.

    What am I missing?

    • The EPA (and some states like California) have requirements that must be met before cars can be sold. So it's not so much an issue that the buyers wouldn't have selected the vehicles because they were more polluting -- it's that the vehicles shouldn't have been available for sale *at* *all*.

      And once they fix the problems, then the fuel efficiency will be lower, which is one of the factors that many buyers consider (and you mentioned yourself).

      Another issue that I haven't heard discussed if the CAFE standar

    • Car owners are upset because they will surely need to have their cars recalled and have the chips replaced, and depending on which way VW makes the tradeoffs, may find that their cars drive badly after replacement or fail inspection after replacement or both. And also that the resale value of their car just plummeted.

  • Once again, listen carefully. You don't deter/avoid/eliminate malicious behaviour like this by creating more stringent testing methods. What you've here is decided to spend more money to create better compliance testing, in a world where those being tested (car makers) can profit by finding better ways to cheat.

    Congrats, you're just going to breed better cheaters.

    And it's obvious why: your playbook is public, theirs is not. They know how you're going to test them. You don't know how they are going to ch

Remember to say hello to your bank teller.

Working...