I think there was a point where regulators said "diesel puts out lung damaging levels of pollutants killing X people a year. We shall either ban diesel engines ... or make manufacturers make diesel engines that don't pollute that badly"
So they set a safe level.
And no manufacturer has been able to achieve the technology to meet that level.
I know some manufacturers claim they can, but honestly that's like a cyclist claiming that they won the Tour De France without drugs. After so many cyclists have been caught (about half since 1990) it's really hard to take the drug free claim seriously - just as it's hard to take the "our diesel engine does it really honestly guv"
It's still an improvement if the large place that burns stuff extracts more energy than tons of small places burning stuff. Not to mention, it's easier to then replace the few large places that burn stuff with something cleaner than having to replace every single car on the road again.
There are lot of parts in the network. No one claimed they would all make a step transition overnight. Part of that is to replace parts of the network which continue to work, while being capable moving the other parts forward.
It’s actually worse than that - they set goals and then manufacturers started lying to meet those goals, which told the regulators that those goals were attainable and so they set new goals. The entire notion of clean diesel is a farce and has been built on lies from the outset.
It's a typical corporate effect. Management, with no expertise or even competency, creates absurd schedules and goals and thinks that somehow, just by fiat and edict, they can bend physics and make it so.
Why is that? Are the parts necessary to meet emissions requirements extra expensive and so are only worth installing on commercial vehicles which are more expensive, vs passenger vehicles have a different profit margin, or is it something else, like commerical vehicles have lower standards for noise or higher standards for maintenance?
Commercial diesel engines run for millions of miles / tens of thousands of hours versus passenger or non commercial is measured in hundreds of thousands of miles typically and thousands of hours. Commercials idle for far longer especially in dense areas like cities, rest stops, distribution centers and ports. Maintaining a fleet of diesels can be written off as a business expense (OPEX) while most non commercial use can’t or isn’t. Would be my guess.
A combination of factors; cost, customer expectations, and convenience.
It's easy to produce a lot of power, it's not easy to do so reliably and within emissions specs. That's where the cost comes in, and where customer expectations come in. If VW is going to be offering a 110 kW 2.0 liter engine, well, Mercedes-Benz can't come in and offer a 90 kW 2.0 liter engine just to meet specs. At the end of the day, margins are fairly thin and regulators are compliant. It's cheaper to just cheat the emissions than make the engine meet emissions specs.
The convenience factor is diesel exhaust fluid (AdBlue); the stuff really does work very well. However, dispensing it at the most effective rate in regards to emissions would mean it has to be topped up between service intervals; very inconvenient. Increasing the tank size is a non-starter because packaging space in modern vehicles is at a premium. So the dirty secret (at least for Mercedes-Benz, confirmed by one of their engineers) is that they calibrate it to last service intervals; not to meet emissions. It's only in rare cases where the owner has to refill the tank themselves.
In regards to the AdBlue situation, if you're in Europe where there are a lot of diesel passenger vehicles and also a lot of diesel trucks and buses, next time you're in the city or on the highway, pay attention to the characteristic diesel stink, either as a pedestrian or a driver. You're never going to smell it from a truck; it'll always be a passenger vehicle. :)
Here in the western US we have em too. Diesel bowsers have a second blue hose attached. You put one in the fuel port, one in the def port, and then let them both fill up
I think I've seen those pumps here in Germany as well, and I'm pretty sure you can get AdBlue in canisters (I have one in the basement that I forgot putting back into a rental).
Yes, this (plus that the whole thing is more convenient than filling up the washer fluid) makes the whole "we don't want the customer to fill it up" even more absurd.
Not surprising, the commercial version of the ISB in question only goes up to 360hp rating, usually they run 300 or 340 hp. The commercial versions all have SCR (Urea Injection) along with DPF an EGR.
In the pickup it's rated 400 hp. I wouldn't be surprised if in the pursuit of HP / Torque to keep up with Ford and Chevy's diesels they cut corners on the emissions. The hotter you run diesel the more powerful it gets, the leaner the more efficient it gets, hot and lean makes NOx in the combustion chamber from oxygen and nitrogen, the more NOX the more SCR you need to reduce it back in the exhaust.
Used to be known as emissions cheating before VW, now it's emission fraud.
Imo every manufacturer does/did it for gas and diesel engines. I've heard of gas cars in the 90s assuming that they're on a test stand if you rolled down the window shortly after starting and kept it down and reduced power output. And stories like that.
I've never heard the bit about windows and I don't see how that wouldn't be very obvious in a pre-canbus car; "why does the ECU have a sense line to the driver's side window?"
What is easy is looking for the car to be started and then held at a raised idle. Once cars had traction/stability control, the ECU could look for mismatched front/rear wheel speed, or the steering wheel not being turned more than a few degrees.
> And no manufacturer has been able to achieve the technology to meet that level.
Urea-injection seems to work (that's what Mercedes does). But it requires another tank and special equipment.
And VW could meet the emissions requirements. That was the cheat. When they detected that they were being tested, they tuned the engine to meet emissions requirements. During normal use, the engine would make better power or efficiency but higher emissions.
It is wild to me that OEM's can straight up include tech to defeat emissions testing and be allowed to continue building engines after that is found. Like, all regulators should operate from the assumption that the entities they're regulating are working in good faith: but once it's been demonstrated they are not, how can you ever trust a product they produce again? They and every other company found to be doing this should be barred from producing engines for good. Or at the very fucking least, be subject to an INCREDIBLE level of scrutiny regarding their software for a solid many years to follow. Like, every single line audited for compliance.
Yet another reason the software running on these automotive systems should be free and open source. It would be great to see the control logic and know that it's doing the right thing.
Random idea. Change emissions requirements based on GPS coordinates. That's sort of what they do in shipping. Ships burn cleaner fuel in port. If a semi is driving around Los Angeles low emissions are important. In middle of Kansas not as much.
That said phasing out diesels is better idea over the medium run.
I think everyone was doing it. After VW got caught, literally every other foreign car maker pulled all diesel models out of the US market. The only manufacturers still selling diesel here are domestic (and maybe MB’s Sprinter?), and that’s only because 10,000 lb GVWR diesel vehicles are allowed to bypass emissions.
Is there any detailed information anyone can find on what they allegedly did, at a technical level? "Defeat device" is such a broad category of term that it's useless for understanding the details of what it's claimed they did.
The Justice.gov writeup [0] isn't any better.
> The company allegedly installed defeat devices on 630,000 model year 2013 to 2019 RAM 2500 and 3500 pickup truck engines. The company also allegedly installed undisclosed auxiliary emission control devices on 330,000 model year 2019 to 2023 RAM 2500 and 3500 pickup truck engines.
I'd be interested in reading technical details on what, exactly, they did or didn't supposedly do.
I want to know the technical details of what Cummins allegedly did "enough that they're not arguing a massive fine, while claiming they didn't do it on purpose."
Claiming they "installed defeat devices" isn't nearly enough detail. How did it alter either the engine combustion cycle or the emissions control system behavior?
Who knows? I believe VW chief engineers got in jail not knowing what their subordinates did. I really tried to understand the VW scandal but I could not.
When operating an engine (gasoline or diesel), you can make a range of decisions about how you run it. Fuel/air ratio, ignition timing, exhaust gas recirculation amounts, boost pressure, etc. There are a lot of ways to accomplish a given power output, and generally some will be more or less efficient in terms of fuel quantity used, and some will be better or worse for emissions. You'll usually find that "optimum for one" is not "optimum for all," and so the ECU (engine computer) has to decide where to run the engine for any given situation.
The (increasingly invalid) assumption made by the EPA test cycles for emissions is that "What it does on the test treadmill will generally reflect what it does on the road." So they've got some (well known) set of drive cycles that reflect a range of driving conditions, they put a car through its paces on the treadmill, and this (a) validates that the car meets the relevant emissions standards, and (b) gets you the EPA-rated miles per gallon. Probably other stuff too, but these are the important ones.
VW violated, quite intentionally, the "behaves the same in the test chamber as on the road" assumption. Every time the car was started, it would assume it was being tested for emissions, until something "impossible in a test chamber" happened (like the steering wheel was turned - you don't do that on rolling road treadmills). At that point, once it was confident it wasn't being tested, it would alter the engine maps to be rather a good bit more fuel efficient - but at the cost of NOx (oxides of nitrogen) emissions being also far higher. So, on the road, the result was that the cars beat their EPA rated fuel economy (which is odd - nobody ever does that, as the real world has things like wind), but also violated the emissions regulations they were certified under (while also putting out less CO2 - because it used less fuel than rated for those conditions).
Most likely they've been caught with memos saying to install a device that would trick past the test. It doesn't really matter what the devices did on an engineering level or even whether they worked at all, and the regulator may not even know or care; the hard part in a case like this tends to be showing criminal intent.
I'm guessing they're doing it during state emissions certification. When they do those tests they hook up to your OBD2 port [1] and generally a tube goes onto your exhaust. It'd be pretty trivial to detect, "Cable is hooked up, exhaust has more back pressure. Tune engine to X mode."
I thought they went away with the tube testing once the OBD data got good enough. Trick to passing the tube tests was to take your vehicle on a nice long ride to get hot (especially the cat).
What I disliked about state emissions testing is that they didn’t do analysis/sampling of data to take an evidence-based approach to target testing against vehicles that tended to fail.
It depends on the vehicle and state. In general, "tube testing" went away on gas cars when OBD-II was standardized - they just ask the car, "Are you happy with your emissions system condition?" and accept the answer, though it's not utterly trivial to forge the answer.
The fundamental problem with all the emissions testing is the gap between "as it's supposed to be done" and "as it's actually done." The people actually doing the work tend to be, more or less by definition, "car people." And they reasonably enough tend to avoid the more "engine abuse" ends of the test spectrum when they can do so - which does reduce the effectiveness of the tests in some cases.
For older diesels (my '97 7.3 Powerstroke, for instance), the "snap test" is supposed to involve flooring the throttle and letting the engine bounce off the rev limiter, while the sniffer sees how much smoke you emit (with a comically high pass limit - something like 40% opacity, which is really "a very dark exhaust plume"). The one and only time my engine has seen the rev limiter in my car was when I emissions tested it shortly after buying it, and did the procedure as written. Every other time I've tested it, someone else has run the test, and they'll bring the engine up quickly (but not floored) into the test RPM range, which doesn't involve going to redline - think 1800-2000 RPM instead of 3500 RPM. They way they do it, you pretty much can't fail if you drove the truck there, unless you've got a check engine light glowing or obviously missing emissions control equipment (no bulge in the exhaust pipe where the cat should be).
Because the people doing the test, very reasonably, don't want to be responsible for older engines throwing rods out the side of the block, and there is zero reason to have a lot of the older diesels at redline, no matter what you're doing (I'll let the engine spin to 3k when I'm pulling a heavy trailer up a long grade, otherwise I never even see 2500 RPM).
And there's also the various "ways around the test." A friend who used to have a rather fun to drive car noted that his sniff tests always resulted in the emissions profile of his mechanic's very clean, stock, Honda Civic.
There's only so much you can do, unfortunately, when people are motivated to go around the edges of the system. And I'm not sure "clamping down massively" accomplishes much at this point.
Engine computers can easily be reprogrammed to make the engine run at all sorts of different operating bands. You could easily detect state emissions equipment, since it must pull information from the ECU. To me- these devices are either pre-programmed operating modes that produce clean emissions but the vehicle would not operate under this tune (emissions testing is often not under load) and so once back on the road, returns to the original operating mode. OR, entirely fake modules created to trick emissions systems without altering the operating mode.
PS check out Megasquirt if you want to learn about DIY Fuel Injection. I ran a home brew Megasquirt/Megaspark system off an IBM laptop in my Volvo 740 Turbo back in the 90s, one of the things that got me into hardware.
Makes sense. As I said, emission testing is often not performed when the vehicle is being driven/under load. States could incorporate dyno testing, but its technical, complicated, and unsafe.
You are correct, the steering wheel approach is very smart... all I was suggesting was that dyno testing would limit the ability to defeat, but testing stations would also need to simulate more of the vehicles systems.
Manufacturers were doing this well before OBD2. They'd look for a condition where the car was started and then idle raised and held raised, for example.
Once things like CANBUS became popular, as well as ECUs getting integrated with ABS/traction control, they'd look for drive wheels spinning and but not undriven wheels. When stability control systems became a thing, they could monitor steering angle and look for the car moving but not being steered, ditto for the accelerometer not detecting acceleration and cornering.
Tuners knew all about this shit because they disassembled ECU code and looked at ROM formats. They could point you right to the table used for when the ECU thought it was undergoing emissions testing. The cheating was plain as day and no government regulators cared enough to look. They probably even knew it was going on.
Here's the billion dollar question: the ECUs aren't developed by the car manufacturers. They aren't even coded by thr manufacturers. The car manufacturer just puts in maps for various engine parameters. The ECUs are supplied by companies like Bosch, Hitachi, Nippon-Denso, Magneti-Marelli, and so on.
They almost certainly knew this was going on and were complicit. Why have none of them been held responsible?
> Tuners knew all about this shit because they disassembled ECU code and looked at ROM formats. They could point you right to the table used for when the ECU thought it was undergoing emissions testing.
I guess the EPAs awards aren’t that great.
I wonder if they also run cars sloppy when it detects something that resembles a drive-cycle for fuel economy testing. e.g. Stop charging battery. Limit A/C or turn on mild heat to reduce radfan load. Limit power steering force. Dim lights 10%. Limit acceleration (gotta be a big one).
Probably enough data out there to identify offenders by comparing real world fuel economy data to official fuel economy.
The fuel tests basically already do all that as protocol. Accesories off. Accelerations are specified and extremely low (I think it was something like 30 seconds to accelerate from 35 to 55).
"A $327.5 million settlement has been reached with Bosch, the supplier of the emissions software that was installed in some 2.0-liter and 3.0-liter Volkswagen, Audi and Porsche diesel vehicles. The Bosch Settlement follows similar settlements with Volkswagen, Audi and Porsche (called the “2.0-liter Settlement” and the “3.0-liter Settlement”, or together, the “VW Settlements”)."
Yup, and I would argue most are not even doing the tuning, Bosch ECUs come pre-mapped to the FI system selected for the application. Maybe some performance models or chassis builders modify in house, but think of all the shared platforms used by all the big and now consolidated automakers...
I mean, I can tell you first hand how easy it is to trick an OBDII emissions test just by messing with the ground wires associated to the O2 sensors and bam, you have different values being report by O2...
Early OBD2 downstream O2 defeats were a simple astable timer (and, for some cars, a power resistor to simulated the heater draw).
The state testing in most states now is ‘make sure the ECU is reporting “READY” and has no emissions-related codes set.’
Since the advent of OBD2, it’s become rare for a state inspection to run the vehicle live for the purpose of emissions testing. Plugging in the OBD2 scanner is way faster, cheaper, more convenient, safer, and easier (albeit slightly) on the car than using a dyno.
Only on older cars, which do not have OBDII. OBDI and before that require a sniffer. OBDII has enough sensors that the sniffer isnt needed. Am in CA, and only my foxbody ever got a sniff test. All of my other (newer) cars, the sniffer wasnt used
It's actually surprising to me just how differently an engine can perform with different tuning. The ECU can control all sorts of parameters depending on the engine. Things like fuel/air mix, ignition timing, valve timing, shutting down some cylinders, skipping ignition cycles, a lot more than you'd ever expect.
You can tune an engine to sound like a braap braap muscle car or have it be nearly silent. You can burn all your fuel to get a few more horses or blow black smoke, or you can tune it to consume no fuel, emit nothing, and produce nearly no torque.
Plenty of people tune their engines with aftermarket kit and/or software. There are even simple plug and play kits of extremely dubious quality any average Joe can install.
What these defeat devices do is detect the testing equipment or conditions and tunes the engine in a way that meets requirements. Once the vehicle is out of testing, it reverts to the factory tuning which usually gives more power at the cost of illegally high emissions.
That's really all it is, there's not too much magic involved. Just simple cheating.
That enforcement action is from 1998. There was even a consent decree but they've been doing this bullshit for a quarter century. It isn't a small isolated incident, it's literally built into the culture of the company. We need a corporate death penalty for repeat offenders like this.
US company so a penalty that is less than the profit, "no wrongdoing admitted" (its a defeat device!), no criminal investigation (there would be precedent), press release goes out on the Friday before Christmas.
I mean, we don’t have any details yet. What if the “defeat device” is something like “if $SENSOR is reading a little high, turn on service light and operate as normal otherwise” As a customer I’d be pissed if my truck wouldn’t run.
Not going to pass an emissions test with a service light on. But I agree that more details are necessary before we can know that the statement is false.
There's no way to make a diesel or gasoline car that's not actively bad for the environment. Electric is the only way. (And maybe hydrogen, but that's quite a ways off)
An electric car is bad for the environment (described generally, not in the laser-focused "CO2 emissions are the only thing that matters!" modern sense) in almost all the same ways an ICE vehicle is, just with a reduction in runtime carbon emissions, and with quite a bit more mining going into the raw materials.
Hydrogen, meanwhile, is nonsensical in every way you care to look at the problem, unless you look at it through the lens of "petrochemical suppliers who want to ensure that a future vehicle fleet needs to fill up at stations they supply with fuel that can be rapidly delivered in a 5 minute window." And maybe shipping, but even there, I think metal-air batteries that are smelted for recharging are likely to work better. And that's before you get into what a devious little pain in the ass hydrogen is to deal with at a chemical/technical level.
> An electric car is bad for the environment (described generally, not in the laser-focused "CO2 emissions are the only thing that matters!" modern sense) in almost all the same ways an ICE vehicle is, just with a reduction in runtime carbon emissions
No it isn't. You can charge them entirely with renewable sources (or nuclear), reducing traditional air pollutants in addition to CO2. This is even true when still using fossil fuel power plants, which emit fewer pollutants per unit energy than small engines.
Electric cars use regenerative braking, reducing the use of friction braking that releases brake dust into the environment.
> and with quite a bit more mining going into the raw materials.
Electric cars are only marginally heavier than gasoline cars because electric motors are light and the batteries take the place of not only the fuel tank but the entire powertrain and emissions equipment. It's not obvious how mining two tons of raw materials is dramatically different than mining 1.9 tons of raw materials, and in any event the raw materials can be recycled from older electric cars once there are older electric cars to recycle.
You're looking at the problem from a purely "atmospheric emissions" perspective. Think about things like "wildlife disruption from a car-centric culture with lots of roads," "microplastics from tires," etc. All of those can very reasonably be considered "environmental effects of cars," if you want to take a broader view of the problems.
Renewable energy sources also have their own impacts on the environment - the mining for raw materials, the cement production used for wind turbines (look up the amount of concrete per MW nameplate - it's quite staggering), etc. One may reasonably argue that these are less important impacts than CO2 emissions, and make one's arguments (I think quite reasonably so), but to round all these to zero for renewables is dishonest, at best. The same goes for the amount of mining required for the raw uranium to be refined into nuclear fuel - quite a bit of earth has to be extracted and processed to get the fuel quantity. Again, I think it's lower impact than fossil fuels, but it's certainly not zero - by quite a large amount. Is "Filling a mountain valley with acidic mine tailings" an "environmental impact"? I'd argue yes, though plenty of people seem to not care, as long as it puts less CO2 on the air. Hence my observation that a lot of people are laser-focused on "CO2 emissions only."
As far as mining goes, EVs require a rather substantially different mix of raw materials than ICEs - so "pounds of earth removed per pound of car" are going to be substantially different. Again, you can argue (and people have, fairly reasonably) that this is a tradeoff worth making, but you can't pretend that "Because the cars weigh about the same, they have the same production impact," when one has a thousand pounds of lithium ion battery pack.
My opinion is that LiFePO4 based PHEVs are the current optimum, but I'm aware I'm in a minority there.
> Think about things like "wildlife disruption from a car-centric culture with lots of roads,"
But now you're making a context-specific argument. It's not about the car, it's about where you put the roads. And then you have to weigh that against alternatives. If you replace cars with buses, you still need roads. Trains need tracks etc.
> "microplastics from tires," etc.
But this isn't any worse than ICE cars. Making some things better and some things the same is a net improvement.
And you could conceivably address this by making tires out of something else regardless of what kind of powertrain you're using.
> One may reasonably argue that these are less important impacts than CO2 emissions, and make one's arguments (I think quite reasonably so), but to round all these to zero for renewables is dishonest, at best.
It doesn't have to be zero if it's less than the cost of oil exploration, production, refining and distribution, which is the status quo.
> The same goes for the amount of mining required for the raw uranium to be refined into nuclear fuel - quite a bit of earth has to be extracted and processed to get the fuel quantity. Again, I think it's lower impact than fossil fuels, but it's certainly not zero - by quite a large amount.
The difference between the energy content per kg of burning oil and fissioning uranium is literally more than a factor of a million. The amount of uranium you need by comparison is basically zero.
> "Because the cars weigh about the same, they have the same production impact," when one has a thousand pounds of lithium ion battery pack.
Lithium battery packs are only a single digit percentage lithium by weight. The lithium isn't a large proportion of the mass of the vehicle. The largest component of the mass of the battery is typically graphite, followed by aluminum.
Modern gasoline vehicles don't really release noxious emissions, at least if they are well maintained. Of course CO2 is not benign, but electric and hydrogen will also have CO2 emissions at this point, just further up the supply chain for the energy.
Is there any article explaining what they actually did, or are alleged to have done? With the VW scandal, there was a fairly in-depth discussion of the technical aspects of it. But all the articles about Cummins seem exceedingly vague.
IIRC, VW had code to detect emissions testing and reduce performance at that time. What did Cummins do?
Air pollution kills 10 million/year[1]. All the legacy carmakers + fossil fuel companies are responsible for this. Nearly all the big car manufactures have done this[2-6], and probably continue to do so, these small fines are not a deterrent, just cost of business.
I was just thinking "wasn't Volkswagen vilified for this not long ago?"
So basically all automakers with diesel do the needful and fake around results anyways. I guess these rules are merely up for interpretation loosely, and this is why the planet is dying.
I have an old Dodge Ram with a Cummins 12 valve engine. The next generation version of my same truck came with a 24 valve engine and more emissions controls. The 24 valve Cummins engine is famous for the ability to re-tune and "delete" features on the engine for better performance and dirtier emissions.
The 12 valve I own only needs a screwdriver to change the fuel mix and increase horsepower. These engines are considered highly desirable because they are very simple with unusual durability (the design comes from Cummins industrial engine product line). When you see the a-holes on YouTube "rolling coal" it is typically from one of these Cummins engines.
These are older than the trucks mentioned in this article, but I wonder if Cummins continued to allow easy adjustments to defeat emissions controls.
The VP44 24v didn't have any more emissions controls than the 12v in the traditional sense. No cat, no DPF, no EGR. It just ran cleaner because of tighter control on injection timing and volume due to the switch from a purely mechanical pump/injectors to electromechanical pump in the VP44, then to fully electronically controlled injectors in the common rail after 2002.
This change to electronic control is what made dial-your-horsepower and emissions skirting like this so trivial.
reply