By on October 8, 2015


A former federal official and the Environmental Protection Agency said that German supplier Bosch didn’t supply Volkswagen — or other automakers — with cheating software, implying that Volkswagen engineers acted alone in deceiving emission tests, Reuters reported (via Automotive News).

According to the report, Bosch supplies the engine control management unit for most four-cylinder diesel passenger cars, including Mercedes-Benz, BMW and others. Both BMW and Mercedes have said their cars do not have software that cheats emission tests.

According to the report, the U.S. House Energy and Commerce Subcommittee on Oversight and Investigations will consider recommending changes to the way regulators test emissions to more closely match real-world conditions.

Bosch said Tuesday that it supplied the units to automakers but “how these components are calibrated and integrated into complete vehicle systems is the responsibility of each automaker,” Reuters reported.

Last month, a report said that Bosch warned Volkswagen of its illegal software in 2007. The supplier didn’t comment on the report, citing contractual privacy with the automaker.

A former EPA official told Reuters that software to detect EPA test cycles would require significant effort by an automaker.

“It is highly unlikely that this additional software is in any computer that does not have a defeat device, as the code requires significant additional resources to write and it would be of no use unless a defeat device was being used,” John German, a former EPA official and a senior fellow at the International Council on Clean Transportation, told Reuters.

Get the latest TTAC e-Newsletter!

16 Comments on “Bosch Didn’t Supply Cheating Software on Dirty Diesels...”

  • avatar

    ““It is highly unlikely that this additional software is in any computer that does not have a defeat device, as the code requires significant additional resources to write and it would be of no use unless a defeat device was being used,” ”

    Is German confused, or have we been misinformed?

    I was under the impression that the “device” was just software in the ECU, not a piece of hardware that did anything physical to the engine/emissions system as such.

    Is there such a physical device, and if so what does it do and where is it?

    If not, German really needs to stop talking about a “device”.

    • 0 avatar

      A defeat “device” can be a piece of software code. They’re saying that the particular bit of code serves no legitmate purpose, and it would have been a rather significant undertaking to implement.

    • 0 avatar
      Lack Thereof

      The “defeat device” is all in software. But the reason the term “defeat device” is still rolling around in the media, is because it appears in the relevant regulations that make it illegal to build a car which operates differently under testing than it does on the road.

      German seems to be referring to two separate functions of the software here. 1) One function which merely detects EPA test cycles, and sets a flag somewhere that indicates the car is being tested. 2) A separate function, which does nothing until the “being tested” flag is set, upon which it activates Good Citizen Mode.

      Basically, German is saying “there’s no need for an ECU to detect if the car us undergoing EPA testing, unless something else in the system is planning on using that information for evil.”

    • 0 avatar

      Based on the language, in software development it means VW engineers reverse engineered a command sequence to detect EPA testing. This would take a lot of trial and error, and it is not the test mode for diagnostic / debugging that Bosch would provide for R&D.

  • avatar

    Sigivald – agreed. I thought it was just software.

  • avatar

    Not sure if I even understand what this guy is saying. He must be a political appointee or bureaucrat. They are experts at talking for hours and saying nothing of any consequence.

    • 0 avatar

      It is engineering language. Bosch provide a test mode for R&D, VW reverse engineered a detect sequence of EPA test cycle (connector detection, command sequence detection, duty cycle detection, whatever) and then switch into the “test mode” for the purpose of fooling a test result.

      Bosch does not provide this detection mechanism so they are not liable.

  • avatar

    Not only is the defeat device not hardware, in VW’s case it’s the absence of hardware that everyone else uses. VW’s Clean Diesel claim was that they didn’t need the urea exhaust treatment system to get all of performance, mileage and low emissions on the 2.0L TDI. Now that turns out to be untrue.

    If they have to add the urea system hardware it’s going to be especially costly. If they do a software only fix then one or both of performance and mileage are going to suffer.

    • 0 avatar
      Lack Thereof

      Well, it’s not completely untrue. They were able to succeed in meeting emissions standards without the urea injection. They were just only able to do it for long enough to pass the test.

      • 0 avatar
        Nick Engineer

        “They were just only able to do it for long enough to pass the test.”

        I don’t know if this is entirely correct. They were able to do it by taking away something from the car that the smog test doesn’t care about: performance, and fuel efficiency. In theory they could use their non-urea system all the time (a lean NOX trap that needs frequent regen cycles which use up diesel fuel, but how that would hold up in a full-time duty cycle on existing cars is anyone’s guess. A urea-based system would be more reliable, of course, but that’s the lonely corner VW has painted themselves into.

        • 0 avatar
          Lack Thereof

          Well, to be fair, the same EPA test used for emissions is also used for fuel economy testing. So whatever fuel economy impacts are seen will likely only bring the MPG of TDI cars back down to their EPA rated economy, rather than regularly exceeding it like they do now.

          But the longevity of the the NOx trap under full-time service with frequent regen cycles will be a definite warranty concern for VW. These warranty costs will probably be the determining factor as to whether VW retrofits these cars with urea systems, or performs a software-only recall.

          • 0 avatar
            Nick Engineer

            “Well, to be fair, the same EPA test used for emissions is also used for fuel economy testing.”

            yes, i stand corrected.

            This would give support to what I suspected for some time now: that VW’s advertised fuel economy on their marketing materials was intentionally based on the emissions controls being fully operational and that a software fix will still shield them from lawsuits about false claims. The fact that customers were getting much better than advertised won’t be nearly as bad in terms of legal exposure.

  • avatar
    Nick Engineer

    John German is correct. All he is saying is that it is very unlikely the defeat code (*) would be included in any ECM that doesn’t need it.

    (*) the defeat code is considered a defeat device by the clean air act. The car maker is required to identify all AECDs whether they are hardware or software.

  • avatar

    I smell complicity and I see posturing.

    We’re gonna need a lot of popcorn before this one gets put to bed.

  • avatar

    There’s an awful lot of space between the automaker is responsible for integration and not supplying the xode which has no other use than cheating a test. Someone hired a lobyist level pr mouthpiece. I will be shocked if this isn’t shared between at least these two firms.

    • 0 avatar

      Maybe not. Volkswagen makes a large number of diesels, but Audi doesn’t, and the few Audi diesels are sold to people who don’t really care about the cost of fuel. Audi wasn’t touting the urea-free aspect of its diesels, just Volkswagen. The whole idea was lower cost diesels for mass production, and sale to people who DO care about fuel mileage, as well as the cost of the vehicle.

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • Corey Lewis: “09 and 11 was the same M35/45” Not at all.
  • Onyxmandingo: From what I read that most likely is the new maxima
  • Onyxmandingo: I have been driving Maximas since the 90’s with my first one which was a rare 95 SPECV version...
  • Steve Biro: Agreed. I think my point and yours are pretty much the same thing.
  • thornmark: NOT owning a VW is a feature, VW is just admitting it

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Mark Baruth
  • Ronnie Schreiber