4 ms·

I don't understand why the emissions procedure was so easy to trick in the first place. Obviously we want to know how much the emissions are in real life drivin

I don't understand why the emissions procedure was so easy to trick in the first place. Obviously we want to know how much the emissions are in real life driving, so why isn't the test simply that we find a representative sample of 100 or so people who have the cars and are using them? Just shove the measuring kit on the cars, maybe pay people for the inconvenience, and see what the numbers come out as? We're not talking about figures that were almost legal here, which maybe could justify a lab test, they were pumping out many times the amount that was allowed.

Oddskar 5y ago
Because you need to have this measured out by the time the car is available for purchase (maybe even way in advance?), not after.
- hef19898 5y ago
  Also comparable standardized test procedures. Those are needed for certification proposes.
- SilasX 5y ago
  Not really, you could just do it as tax proportional to how much in-practice the company’s cars pollute. Though that would introduce a greater difficulty in risk and cost management since it’s harder to predict and control in advance.
maeln 5y ago
Tests needs to be standardized to ensure that they are replicable. If running the test several time wouldn't give the same result (within a small tolerance margin) manufacturer would cry unfairness and they would be right. What if in one run more driver use the freeway and in another where stuck in cities traffic ?
- lordnacho 5y ago
  I tried to address this in my comment. If if was some marginal thing where there was a chance they were close, then maybe it would make sense to have a standardized lab test. If someone is 50kg overweight we don't need a special procedure with a sensitive scale to decide that.
bagacrap 5y ago
I'd assume the VW engineers could still create a defeat device (ie detect the test equipment and alter performance) for that situation
bri3d 5y ago
This is pretty much how the defeat was discovered - by a group who were curious about how US diesels could be so great compared to their European counterparts, using sniffers attached to vehicles performing normal road driving. Unfortunately, this wouldn't work very well as a certification test, as it would require standardization. But, as a failsafe in addition to a standardized test plan, I think this would be great - basically "we'll give you the procedural test you know and love, and we'll drive 100 normal road miles with more lenient thresholds in place. Fail either and you don't get a certification."
r00fus 5y ago
Likely the emissions procedures were developed with the "help" of industry (VW included) lobbyists as well. Any remediation from Dieselgate that doesn't include structural changes like improved testing means the auto industry has won.

Oddskar 5y ago

Because you need to have this measured out by the time the car is available for purchase (maybe even way in advance?), not after.

hef19898 5y ago
Also comparable standardized test procedures. Those are needed for certification proposes.
SilasX 5y ago
Not really, you could just do it as tax proportional to how much in-practice the company’s cars pollute. Though that would introduce a greater difficulty in risk and cost management since it’s harder to predict and control in advance.

maeln 5y ago

Tests needs to be standardized to ensure that they are replicable. If running the test several time wouldn't give the same result (within a small tolerance margin) manufacturer would cry unfairness and they would be right. What if in one run more driver use the freeway and in another where stuck in cities traffic ?

lordnacho 5y ago
I tried to address this in my comment. If if was some marginal thing where there was a chance they were close, then maybe it would make sense to have a standardized lab test. If someone is 50kg overweight we don't need a special procedure with a sensitive scale to decide that.

bagacrap 5y ago

I'd assume the VW engineers could still create a defeat device (ie detect the test equipment and alter performance) for that situation

bri3d 5y ago

This is pretty much how the defeat was discovered - by a group who were curious about how US diesels could be so great compared to their European counterparts, using sniffers attached to vehicles performing normal road driving. Unfortunately, this wouldn't work very well as a certification test, as it would require standardization. But, as a failsafe in addition to a standardized test plan, I think this would be great - basically "we'll give you the procedural test you know and love, and we'll drive 100 normal road miles with more lenient thresholds in place. Fail either and you don't get a certification."

r00fus 5y ago

Likely the emissions procedures were developed with the "help" of industry (VW included) lobbyists as well. Any remediation from Dieselgate that doesn't include structural changes like improved testing means the auto industry has won.