After the U.S. Environmental Protection Agency today accused Fiat Chrysler of having “defeat devices” for emissions in diesel Jeep and Ram vehicles, company boss was adamant that Fiat Chrysler did not cheat like Volkswagen, calling such a claim “hogwash.” That may be a bit far, but after carefully examining everything, I’m not sure it’s apples-to-apples with VW’s Dieselgate, either.
Despite the fact that we still don’t know much about the new allegations, there are already clear differences between this and how Volkswagen cheated—the most obvious being that the EPA straight-up announced that VW was cheating, and eventually that the cars would have to be bought back or repaired. In the case of Fiat Chrysler, the EPA is actually giving the company the opportunity to explain itself.
Another obvious difference is scale. Here we’re talking about 100,000 Ram and Jeep vehicles, while VW cheated with nearly half a million four-cylinder cars in the U.S. and millions more globally. But to really figure out how they’re different, we have to dive into what the EPA alleges in both cases.
Let’s get straight to what Volkswagen did, because it was absolutely egregious, involved a cover-up, and was very, very illegal. The company installed “defeat devices” on nearly 500,000 cars to trick regulators into thinking the vehicles were cleaner than they really were in real-world driving. Defeat devices are defined by the EPA as:
...an auxiliary emission control device (AECD) that reduces the effectiveness of the emission control system under conditions which may reasonably be expected to be encountered in normal vehicle operation and use.
Keep in mind software can qualify as a “device” under this definition. It’s just a legal phrase.
EPA tests found that VW’s devices actually sensed when the vehicles in question were on the emissions rolls (dynos) being tested, and if they were, the software adjusted certain emissions parameters to clean up NOx from the tailpipe.
Volkswagen’s device used wheel speed and steering wheel sensors to tell when the car was being driven by a human on a road, or if it was strapped to a dynamometer (you can read about how the EPA traditionally runs its testing here), ensuring low NOx when under testing, but not when the consumer drives the car on the same drive cycle.
I emphasize that last bit because that is arguably the most damning part about what VW did: the company literally made it so that pollution during road driving was worse than when the car was tested even if the car was driven exactly the same way. VW played the EPA for fools.
The EPA has yet to explicitly accuse Fiat Chrysler of installing a defeat device that actually senses when the car is on the emissions rolls being tested, and the company’s CEO Sergio Marchionne addressed this in a teleconference earlier today, saying:
There is nothing in the current calibration of the Ram 1500 or the Grand Cherokee diesel that distinguishes between a test cycle and normal driving conditions. This is a huge difference because there has never been an intention on part of FCA to create conditions that are designed to defeat the testing process. That is absolutely nonsense.
But what the agency has accused Fiat Chrysler of doing is failing to disclose all software features that could be considered auxiliary emissions control devices, or AECDs, which are defined as:
...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.
These AECDs, the EPA alleges, cause the 103,828 Jeep and Ram vehicles equipped with 3.0-liter diesels to emit more pollution in conditions “reasonably expected to be encountered in normal vehicle operation and use” than under EPA test conditions.
According to the EPA, Fiat Chrysler allegedly failed to disclose the following eight AECDs, which tamper with exhaust gas recirculation, EGR, and selective catalytic reduction, SCR, under certain vehicle conditions:
The EGR systems in these Jeep and Ram vehicles recycle exhaust gases back into the cylinders to reduce combustion temperatures and lower NOx emissions, and the SCR system adds urea to the exhaust stream to convert NOx into nitrogen and water.
In other words, these are the two systems responsible for lowering a vehicle’s NOx output under certain conditions. Software adjustments to these two systems work have a huge effect on tailpipe emissions, and the EPA thinks Fiat Chrysler is fiddling with these devices illegally.
The EPA’s notice includes a number of examples where the emissions controls system effectiveness was intentionally reduced by one or more AECD. The examples are meant to communicate that the devices seem to cause the cars to “perform differently when the vehicle is being tested for compliance with the EPA emission standards using the Federal emission test procedure, than in normal operation and use.”
I interpret this as meaning: the vehicles are clean when driven under the conditions described in the EPA’s federal procedures, but under other driving conditions considered “reasonable,” the cars produce significantly more NOx.
Here’s where I’m going to get really technical. Examples of how the control devices cause the vehicles’ emissions systems to underperform and are described below:
Example 1: The first example in the EPA’s Notification of Violation claims that AECD #3 above, along with AECDs #7 or 8, increases vehicle emissions by disabling EGR without dialing up the SCR to compensate. The EPA says this doesn’t appear to be done in order to protect the vehicle, and that it yields reduced emissions control system effectiveness during “conditions reasonably expected to be encountered in normal vehicle operation or use.”
Example 2: The second example essentially says that AECD 5 and 6, EGR reduction due to engine temperature and SCR shutoff for catalyst warmup, yield high NOx emissions “at temperatures outside of those found in the Federal emissions test procedure,” but not needed to protect the vehicle.
Example 3: The third example the EPA gives is how AECD #4, when combined with AECD #8, increase NOx under “reasonable” conditions, and that AECD #1, 2 and 5 can also increase the frequency of the NOx-spewing AECD #4.
Example 4: The final example given talks about how AECD #7 and 8 increase NOx “particularly in variable grand and high load conditions.”
What does all of that mean? It means these “devices” cause increased engine emissions in certain conditions that the EPA deemed reasonable, and that may or may not have been encompassed by the EPA’s test procedures.
If you look at example number two above, you see that the EPA is unhappy about the affected Jeep and Ram models producing too much NOx at “temperatures outside of those found in the Federal emissions test procedure.”
So that prompts the question: isn’t Fiat Chrysler only responsible for meeting emissions requirements at the temperatures specified by the EPA?
In other words, if the EPA’s tests go from 20 Fahrenheit to 95 Fahrenheit, should engineers have to design the emissions system to work well at -10F or 0F, temperatures where engineers might want to make changes to the combustion strategy for faster warmup (for example).
Is it not the job of the EPA’s test cycles to provide bookends for automotive design? If the EPA is unhappy with Fiat Chrysler’s emissions systems only performing well in conditions bounded by federal test procedures, is it the EPA’s job to change the spec?
If you read the definition of Defeat Device, it says any device that increases emissions in “conditions which may reasonably be expected to be encountered in normal vehicle operation and use.” This makes it seem like the Federal Test Procedures are really only a guideline, and that engineers shouldn’t design their cars solely to meet them.
If that’s the case, should the EPA provide bounds for what is considered a “reasonable” use case? After all, there are an infinite number of temperatures, grades, and drive cycle that engineers could optimize their systems for.
I’ve reached out to the EPA for further clarification on this, because perhaps I’m missing something here, but they have not gotten back to me. But I can’t help but wonder if this is a case of engineers designing exactly to the EPA’s federal test requirements.
I emailed Fiat Chrysler’s PR person in charge of regulatory subjects, and he basically told me to look at the exceptions enumerated in the EPA’s definition of “defeat device.” Those exceptions include:
(1) Such conditions are substantially included in the applicable Federal emission test procedure for heavy-duty vehicles and heavy-duty engines described in subpart N of this part
(2) The need for the AECD is justified in terms of protecting the vehicle against damage or accident;
(3) The AECD does not go beyond the requirements of engine starting; or
(4) The AECD applies only for engines that will be installed in emergency vehicles, and the need is justified in terms of preventing the engine from losing speed, torque, or power due abnormal conditions of the emission control system, or in terms of preventing such abnormal conditions from occurring, during operation related to emergency response. Examples of such abnormal conditions may include excessive exhaust backpressure from an overloaded particulate trap, and running out of diesel exhaust fluid for engines that rely on urea-based selective catalytic reduction.
In particular, FCA’s representative highlighted bullet point number two, suggesting that perhaps these AECDs are in place in order to protect the vehicle against damage, and CEO Sergio Marchionne echoed this in his teleconference today, saying:
If there’s anything at all that — that we may have a difference of opinion on as to what we have completely disclosed - the protection mechanisms that are allowed under the regulations to protect engines under particular circumstances. This is a different type of analysis than the VW story, and I think the sooner we get to that conclusion the better we will all be.
So maybe this whole situation has more to do with interpretation of requirements than with intentional deceit, as was the case with Volkswagen.
First off, Fiat Chrysler not disclosing these emissions control devices is already a violation of the Clean Air Act, so the EPA claims. If so, they’ve already screwed up.
Whether they’ve screwed up as badly as Volkswagen remains to be seen, but I think that, if Fiat Chrysler’s software doesn’t actually sense when the vehicle is being tested, then this accusation isn’t nearly as egregious as Volkswagen’s. Because with Jeeps and Rams, it appears that, if you drive them as specified by the Federal Test Procedures, your emissions will meet the spec; the same cannot be said about VW.
That said, I don’t think this is just a misunderstanding of requirements; if it were, you’d expect Fiat Chrysler to have a better excuse than just a bunch of belligerent quotes from their CEO.
We’ll update this as we learn more from the EPA and FCA in the coming days.
Correction: This article originally featured a photo of a Ram Heavy Duty truck, which has been replaced with a Ram 1500.