Cheating TDI EA189 pollution Volkswagen (seat, audi ...)

Transport and new transport: energy, pollution, engine innovations, concept car, hybrid vehicles, prototypes, pollution control, emission standards, tax. not individual transport modes: transport, organization, carsharing or carpooling. Transport without or with less oil.
User avatar
Grelinette
Econologue expert
Econologue expert
posts: 2007
Registration: 27/08/08, 15:42
Location: Provence
x 272

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Grelinette » 17/05/17, 20:07

Flytox wrote:Start at the top left podcast:
https://www.franceinter.fr/embed/player ... aad0ac3776"width =" 100% "height =" 64 "frameborder =" 0 "scrolling =" no "


Some important passages from the Flytox podcast:
"Legal information has been opened against the Renault brand"
"Renault (and Fiat) diesels are among the most polluting of European cars"
"Some Renault emit more than 10 times the regulatory limit of NOx"
"Renault cars pollute more than WW cars by which the scandal happened"
"This manufacturer is very, very late in dealing with pollution from its cars"

etc ...
1 x
Project of the horse-drawn-hybrid - The project econology
"The search for progress does not exclude the love of tradition"
Christophe
Moderator
Moderator
posts: 79323
Registration: 10/02/03, 14:06
Location: Greenhouse planet
x 11043

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Christophe » 17/05/17, 20:09

All this had been predicted by Bibi here a 18 months: new-transport / volkswagate-yes-but-and-the-other-vendor self-t14222.html
1 x
moinsdewatt
Econologue expert
Econologue expert
posts: 5111
Registration: 28/09/09, 17:35
Location: Isére
x 554

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by moinsdewatt » 27/05/17, 11:30

Volkswagen Scandal: Found Cheating Software Source Code

the 26.05.2017

German and American computer scientists have managed to retrieve the source code of the fake software used by Volkswagen to defraud on its pollutant emissions.

DIESELGATE. We call it the Volkswagen affair: the scandal that hit the manufacturer following its fraudulent polluting emissions caused a lot of noise in 2015. And there is something new, since a team of German and American researchers from the University of the Ruhr in Bochum and the University of California presented the results of an extensive analysis work, which reconstructs the faulty source code, at an international symposium on security. And the results confirm the thesis of massive fraud: out of 900 different embedded firmware, used between 2009 and 2016, more than 400 would have been faked, thanks to a few well-placed lines of code. And Volkswagen is not the only one: Fiat is also pinned, when we learn on May 26, 2017 that a "class action" driven in the United States blames General Motors for the same case.

The video below recalls the principle of fraud, which is already known: the on-board computer detects the test phases and activates the anti-pollution device ... which remains off during normal operation on the road. Result: emissions of nitrogen oxides up to 40 times the maximum allowed by American and European regulators.

The tricked code retrieved by reverse engineering
The known principle remained to identify precisely the portion of the code causing the problem, which took a year. "We found the murder weapon and how the builders used it," said Kirill Levchenko, a computer scientist at the University of San Diego in California. "In fact, the evidence of the fraud was there for all to see." And for good reason: all they had to do was download the software binaries (the manufacturer makes them available to mechanics and the expert public) in order to find the disputed code, by reverse-engineering from the documentation available on the manufacturer's site. .

Quickly, the researchers identified a piece of code called "acoustic condition". At first glance, it seems like a simple engine noise check test. But a closer look quickly reveals that this keyword is in fact used for conditional testing to find out, for the on-board computer, whether the vehicle is going through a homologation test or not. And so, whether or not he should activate his anti-pollution mode. And the fraud is perfected, "one of the most complex special effects in the history of the automobile", argues Kirill Levchenko. Indeed, the system takes into account no less than 10 specific profiles (vehicle speed as a function of time) commonly used for homologation. Another problem: the definition of homologation tests, standardized, is obviously public. What to simplify the cheating, according to the researchers.

Risks of generalized fraud

The researchers also discovered a second variation of the rigged code in the Fiat 500X engine firmware. This is more rudimentary since it is a simple stopwatch, which activates the anti-pollution mode only during the first 26 minutes and 40 seconds of operation, which corresponds to the average duration of a check. "We are the first to publicly identify this rigged device," write the authors.

BOSCH. What is the commonality between Fiat's and Volkswagen's electronics? The manufacturer Bosch, which manufactures EDC17 electronic blocks for diesel engines (see illustration in picture). "We have found strong evidence which shows that the two rigged devices were created by Bosch and then activated by the two manufacturers for their respective vehicles.". Today, the most used method for homologation remains the dynamometer, a mechanic's chassis allowing the immobilization of the vehicle and the free rotation of the wheels, enough to carry out all kinds of tests freely. For Kirill Levchenko, "this type of approval is no longer sufficient, the regulators must also be interested in the conformity of the software". A vast project ...

Readers who are eager for technical details will find below the recording of the conference given by Moritz Contag, co-author of this study.


https://www.sciencesetavenir.fr/high-te ... che_113222
0 x
User avatar
Remundo
Moderator
Moderator
posts: 16126
Registration: 15/10/07, 16:05
Location: Clermont Ferrand
x 5241

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Remundo » 27/05/17, 19:22

Then BOSCH would be as wet, if not more than VW or Fiat ... who would only have exploited the possibilities offered by Bosch. :?

Bosch is doing well.
0 x
Image
User avatar
Grelinette
Econologue expert
Econologue expert
posts: 2007
Registration: 27/08/08, 15:42
Location: Provence
x 272

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Grelinette » 28/05/17, 20:55

Remundo wrote:Then BOSCH would be as wet, if not more than VW or Fiat ... who would only have exploited the possibilities offered by Bosch. :?
Bosch is doing well.

It is probably a consortium of powerful automakers who have collectively decided this fraud, much like secret and illicit trade deals to manipulate a market (eg. The illegal trade agreement between Orange Sfr and Bouygues To agree artificially high tariffs).

What seems surprising to me is that such a cheating involves necessarily many interveners and at all levels, and there are necessarily leaks or denunciations!
1 x
Project of the horse-drawn-hybrid - The project econology
"The search for progress does not exclude the love of tradition"
User avatar
Macro
Econologue expert
Econologue expert
posts: 6515
Registration: 04/12/08, 14:34
x 1637

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Macro » 29/05/17, 09:43

Grelinette wrote:
What seems surprising to me is that such a cheating involves necessarily many interveners and at all levels, and there are necessarily leaks or denunciations!


Bof ... When you are lucky enough to have a good job properly paid as a software developer at Bosch ... You do what you ask, you close it and you pocket the € ... In addition you have to To work on so restricted areas of the software .... That you must be far from knowing what they will serve in finé ... For trying to tinker a can (sleepless nights for several weeks) on the binary of 512Ko Of my alfa of 2000 (calculator edc15) I can guarantee that it already makes a big packet of mappings ... The binary of the EDC17 (of 2010) of the passat has Mamita is only (of memory) ... 1000 times bigger : Shock:
1 x
The only thing safe in the future. It is that there may chance that it conforms to our expectations ...
User avatar
Remundo
Moderator
Moderator
posts: 16126
Registration: 15/10/07, 16:05
Location: Clermont Ferrand
x 5241

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Remundo » 29/05/17, 13:03

absolutely, it is also one of the well-known techniques of military secrecy: splitting up the tasks as much as possible so that no executor has an idea of ​​the overall project. Only Bosch's IT "top managers" knew how all modules and source codes could work together and what their potential was.

So was the design deliberately fraud-oriented, or is it a posteriori That some brains have changed the zinzin somewhat to defraud ... In a software, 2 or 3 parameters, or just 2 or 3 lines of conditional codes to some sensors / set of measurements can radically change the control.

In any case there was fraud.
0 x
Image
User avatar
Macro
Econologue expert
Econologue expert
posts: 6515
Registration: 04/12/08, 14:34
x 1637

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Macro » 29/05/17, 14:54

When I see what my 156 can do by changing only 5 values ​​on the 9 of the torque limitation .... After I do not even talk about the changes that were made on, the advances, the injection times, Pressure limiters atmo, sural, and the small valve that goes well to lure the wastegate to push 1.4bar instead of 1 .... After ... It must be played steam locomotive driver .... Eyes Riveted on various mano and gauges : Mrgreen: Marti when the red log is going to light up you'll feel a push for a few seconds : Mrgreen: : Mrgreen: : Mrgreen: : Mrgreen: But boudiou ... I can guarantee you of a thing .... One finds with a turbo diesel 2000 years .... The sensations we had with a turbo gasoline foldingue years 80 ... Except that. ... It does not explode at the end of 20km (20 000 done with my 156 shiptunnée and it still works) .... And again ... I insisted on staying on the clear side of the force .... By passing from the dark side (exhaust fumes) .... Y'a again (according to my shiptuner) of the couple and the Pmax to gain .... Level consumption ... Those who tell you it consumes less. ... I would say ... Well no ... If you ask the same thing ... It consumes like ... If you ask more ... Effectively cons consumes less (-20% with heavily loaded trailer on highway in my case) ...
0 x
The only thing safe in the future. It is that there may chance that it conforms to our expectations ...
User avatar
Flytox
Moderator
Moderator
posts: 14141
Registration: 13/02/07, 22:38
Location: Bayonne
x 839

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Flytox » 13/09/17, 22:50

1 x
Reason is the madness of the strongest. The reason for the less strong it is madness.
[Eugène Ionesco]
http://www.editions-harmattan.fr/index. ... te&no=4132
User avatar
Gaston
Econologue expert
Econologue expert
posts: 1910
Registration: 04/10/10, 11:37
x 88

Re: Cheating EA189 pollution TDI Volkswagen (seat, audi ...)




by Gaston » 14/09/17, 10:46

It seems to me that right from the start of the case, Bosch said that it was the author of this software version at the request of VW, but that this version was intended for engine testing (in engineering offices) , not to be installed in vehicles.
There is nothing illegal to build a motor for a test bench that does not meet emission standards.

It would be VW that would have decided to install this software in production when they found that they could not achieve the objectives of pollution at the same time as the objectives of cost ...
0 x

Back to "New transport: innovations, engines, pollution, technologies, policies, organization ..."

Who is online ?

Users browsing this forum : No registered users and 134 guests