A closer look at the West Virginia report at the center of the VW emissions scandal.
Volkswagen’s dirty little secret got out last week: the auto giant rigged its diesel cars with a defeat device that reduced pollution during official emissions testing but relaxed on real roads to improve driving performance. But the details at the heart of the news have been openly available since May 2014. That’s when West Virginia researchers reported the results of tests exposing the high rates of NOx emissions in two VW models.
Carder said he's surprised to see such a hullabaloo now, because his team's findings were made public nearly a year and a half ago.
"We actually presented this data in a public forum and were actually questioned by Volkswagen," said Carder.
The enormity of the scandal (in every sense of that word) merits a closer look at the 117-page study at its core that went overlooked for a year and a half too long.
What they did
The International Council on Clean Transportation approached the West Virginia Center for Alternative Fuels, Engines and Emissions to study emissions from three diesel cars under a variety of U.S. driving conditions. The study team’s principal investigator, Gregory Thompson, was joined by Carder as well as Marc Besch, Arvind Thiruvengadam, and Hemanth Kappanna.
The car models were kept anonymous in the paper but we now know their identities. “Vehicle A” was a 2012 VW Jetta. “Vehicle B” was a 2013 VW Passat. And “Vehicle C” was a BMW X5.
The cars went for five test spins in and around three California cities:
- Route 1: Some 43 miles of weekday, non-rush hour highway driving in Los Angeles, at average speeds of 48 mph.
- Route 2: More than 15 miles of weekday driving in downtown L.A., at average speeds of 15 mph.
- Route 3: Nearly 37 miles of rural up and down driving in the L.A. foothills, at 32 mph.
- Route 4: A trip through downtown San Diego, similar in length and speed to Route 2.
- Route 5: A trip through downtown San Francisco, ditto.
Vehicle B—the Passat—also went on an extended road trip of nearly 4,000 miles from L.A. to Seattle.
The cars were warmed up before each test, which is important because it’s normal for a vehicle to spew higher NOx emissions when the engine first gets going. They were also relatively new, none exceeding 16,000 miles, meaning there’s no reason to think the emissions filters would have deteriorated from age. Last, they were outfitted with an emissions analysis system that even the most dimwitted individual with a doctorate in environmental engineering could have designed:
And there was one more critical set of trials: emissions testing at the California Air Resources Board’s vehicle certification facility in El Monte.
What they found
By now everyone knows what happened with those fateful road tests: VW failed spectacularly. NOx emissions for “Vehicle A” (the Jetta) were 15 to 35 times higher than the EPA standards, and those for “Vehicle B” (the Passat) were 5 to 20 times higher. (The BMW passed the test, though its NOx emissions did creep up a bit during hilly driving.)
Here’s the most damning of the report’s many detailed charts (the green line at the bottom reflects the EPA standards):
A trio of other charts is equally revealing. Here the red line at the left reflects EPA standards. For Vehicle A (the Jetta), none of the routes met regulation NOx emissions levels (top). The case wasn’t much different for Vehicle B (the Passat), though one of the highway tests hit the EPA mark for part of the trip (middle). Contrast that with the rule-abiding BMW (bottom)—all of the lines cluster around the EPA standards, save for NOx emissions during hilly trials.
During the long-distance road trip, Passat NOx emissions exceeded the EPA standard by roughly a factor of six, with city driving generally the greatest offender.
For the record, CO2 emissions were below regulation levels for all three vehicles.
What it means
On their own, the high NOx levels of the VW models might have warranted some fines or negative attention for the company. But it was the comparison of NOx produced during road tests with the levels found during lab tests that produced the smoking gun of VW’s defeat device.
The lab tests—when those devices effectively told the cars to behave—produced NOx emissions 50 and 64 percent below EPA regulations for Vehicle (the Jetta) A and Vehicle B (the Passat), respectively. Those findings are shown in the chart below as FTP-75:
Michigan mechanical engineer Anna Stefanopolou explained to Wired just how the cheat device worked. In simple terms, it knew that if the steering wheel wasn’t moving, then nobody was really driving:
Michigan’s Stefanopolou says computer sensors monitored the steering column. Under normal driving conditions, the column oscillates as the driver negotiates turns. But during emissions testing, the wheels of the car move, but the steering wheel doesn’t. That seems to have have been the signal for the “defeat device” to turn the catalytic scrubber up to full power, allowing the car to pass the test.
In their report, the West Virginia researchers don’t make the accusation explicit, but the subtext of their interpretation of the conflicting road-lab NOx results is clear enough:
This is further confirmation that Vehicles A and B were operating as intended and did not have any malfunctions.
Indeed, the cars were operating exactly as VW intended.