By on September 9, 2013


Reports out of the Berlin desk of Reuters suggest that VW could have a fairly large problem on their hands, one that TTAC discussed during heated battles over modular kit architecturesthat of “cascading failures”.

Reports by German auto publication Auto Bild suggest that improperly installed HVAC draining tubes in the all new Mark VII VW Golf can leak water into the driver’s footwell. VW confirmed the problem to Reuters, and suggested that 46 individual cases are known to VW.

On the other hand, Auto Bild has suggested that the number could be as high as 300,000 cars, including the Seat Leon and Audi A3, due to the “cascading failure” phenomenon. The new Golf is built on Volkswagen’s MQB architecture, along with models like the Leon and A3, with the architecture having a high degree of common parts across Volkswagen’s entire range. While this allows for cost savings, manufacturing flexibility and shorter assembly times, it also increases the chances of significant failures, in the event of a bad batch of parts or a common engineering defect. The swift felling of hundreds of thousands of cars (or more) due to the failure of a single component is not necessarily new in the auto industry (we’ve seen mega recalls before) but it would be relatively new for it to affect a group of vehicles derived from a single common architecture.

If this is indeed related to a batch of faulty parts, then it would be evidence of a cascading failure. However, it could also be a procedural error committed by a worker, a poorly implemented assembly process or some combination of the above. This could be one of the more interesting stories of the year, as it may be one of the biggest examples of the downsides of the move to modular “kits”.


Get the latest TTAC e-Newsletter!

39 Comments on “Are We About To Witness MQB’s First “Cascading Failure”?...”

  • avatar
    Extra Credit

    Really? Really?
    A little late to the party with this “news”.

  • avatar

    Come back when you’ve got something interesting like engine mount failures.

    One part failing millions of times has already happened with many manufactures putting the same engine in many vehicles. A good example would be the cruise control solenoid issues with Ford 4.6 and 5.4 V8s for their trucks. That was a recall that got close to a million affected vehicles.

  • avatar

    not sur eyou could blame that on MQB. Those 3 cars already were the same platform, before MQB was invented.

    and if it becasue of faulty parts, why is that a MQB failure? Parts were never faulty before MQB? In addition, the parts stil come from different vendors. Just because it is the same type of part, doesn’t mean only one supplier or one single factory makes them for the world. Even witht he same suppilier you get different qualities in each factory, in each country.

    the Jetta V sure got the same type of part in europe and Mexico, for sure not the very same supplier.

  • avatar

    On one side, with common architecture and VW, I’d be more worried about a true design flaw that can’t be fixed easily instead of a drain leak. Not that VW would ever poorly design anything like exploding panoramic sunroofs, undersized torque-to-yield subframe bolts, carbon collecting intake manifolds, stripping window regulators, malfunctioning coil packs, self-destructing diesel fuel pumps, or snapping camshafts. Becuase hey, that would be totally implausible.

    However, and where things can get really interesting, what happens when someone OTHER than VW decides that there’s a design flaw, like the NHSTA, who have been making some interesting precedents, as of late. Say, they decide a WHOLE PLATFORM doesn’t meet crash standards after approval? Not that that would ever happen, either.

  • avatar


    In paragraph two you stated that it may be due to “improperly install(ed) HVAC draining tubes”. What lacks clarification is if that is specific to the subassembly itself or to the way that assembly line workers are installing the subassembly.

    Thankfully you mentioned this in your piece, but until there is some actual information about the exact nature of this problem (component design or assembly fault) it’s too early to tell if this is something that will affect all HVAC units in all MQB product, or just off of one assembly line/batch of product.

    Still, this is no doubt the kind of thing that keeps managers awake at night.

  • avatar

    Is “cascading failures” a phrase that is being used widely in the press to describe a quality/design problem that impacts many models, or is this nomenclature specific to TTAC?

    The reason I as is that “cascade failure” has a fairly specific definition in engineering – a failure that directly results in additional failures that would probably not have occurred in the absence of the original fault. For example, a timing belt failure which then cascades into mechanically damaged valves and pistons.

    In this case, if the improperly routed HVAC drain dumped water on an electronic component that then failed due to water intrusion *that* would be a cascade failure.

    I don’t disagree with the analysis but I recommend you choose another name for the widespread failures due to common architecture phenomenon, as it is explicitly not a true “cascade failure”.


    • 0 avatar
      Jeff Waingrow

      I had the same reaction to that phrase. Derek, you need to be more precise in pieces like this. “Cascade” implies something altogether different from what you report here. TTAC might be better served if fuller research (and editing) were done before issuance, though I recognize the need to find new material to fill the pages every day. BTW, it seems to me that it would be much easier to have good quality control when you have the smallest number of parts to take account of, which suggests the opposite of what you at least imply here. However, I claim no expertise on this subject. I do know what “cascade ” means, though.

      • 0 avatar

        I’m not an engineer and I understood what Derek was attempting to say, so it’s not a complete breakdown in communications.

        Do we have another word or phrase for this? “Amplified Failure Exposure Potential?”

        • 0 avatar

          I was unaware of the engineering definition of the term but I have used it in the past and nobody objected. Is there a more accurate term we can use?

          • 0 avatar

            How about “Shared Platform Failure/Defect”
            or “Common Architecture Shared defect”?


          • 0 avatar

            There is no widely-recognized definition for this that I know of, but I’d refer to the worst case situation described in these articles as a “platform-level failure” rather than a “cascade failure”.

            It’s important to note that even for vehicles that share their underlying architecture there are lots of internal firebreaks that would prevent a problem from being endemic to every product built on the platform.

            Even with something as widespread as MQB is intended to be there will be variation from manufacturing location, sales region, subassembly or part supplier, and deliberate design changes from model to model or build to build. Every one of these represents an opportunity to divide a problem and greatly reduce the manufacturer’s exposure. Remember the Toyota accelerator pedals that were built for the same application, but only the CTS-made ones had a problem?

            I’d submit that the greater risk of widespread reliability problems comes from smaller parts or assemblies that are used across multiple platforms or even OEMs.

          • 0 avatar


          • 0 avatar
            Athos Nobile

            Pandemic, epidemic, widespread?

      • 0 avatar

        Full research and editing, that’s the ticket. I wouldn’t normally point out a simple sloppy mistake, but since this post is about quality control, I’ll remind the author that there’s no such cars as “the all new Mark VII VW Gold.”

    • 0 avatar

      > In this case, if the improperly routed HVAC drain dumped
      > water on an electronic component that then failed due to
      > water intrusion *that* would be a cascade failure.

      This specific issue was common in the B5 (or was it B5.5?) Passat era. Drain tubes would clog, back up into passenger compartment and soak the “comfort/convenience module”.

  • avatar

    A VW Chief Engineer responds: “Drain tubes installed improperly? Impossible. Ha. Stupid customers who don’t realize our superiority.”

  • avatar

    I’ve had 3 Audi vehicles over the years. Over the years, 3 Audi vehicles had drain tube problems.

  • avatar

    A drain tube, seriously? That is your big news and the end of MQB? If commentators can’t find anything better to carp about, Volkswagen reliablity must be AWESOME. In my day WVs would not run at all, or burst into flames when they did. And now we hear about HVAC DRAIN HORRORS.

  • avatar

    We had 2 Volvo 244 models, both of which drooled cold water into the passenger’s footwell. Not a unique or recent phenomenon.

  • avatar

    I’m guessing the scope of the recall depends if the drain tube lies within the “fixed” zone of MQB, which is between the front axle and the firewall. As I understand it, that is where the vast majority of component standardization is taking place.

  • avatar

    I know two people with late 2000’s A4s whose drain tubes clogged which lead to water getting into the computer and frying it.

    • 0 avatar

      That’s ugly. The lowly Saturn Ion has drain tube issues, but they even had a helpful indent on the box to know where to drill a finger sized hole to clean out the gunk and a weep hole at 2/3 up to let the built up water dribble onto the carpet rather than the blower motor.

      The worst damage of a recall is letting Audi owners know their cars share a lot of parts with VW, SEAT and probably Skoda.

    • 0 avatar

      I’m questioning how that could happen. I checked the routing of the tubing in two available vehicles (not VW: a Chrysler and Toyota). Both are entirely on the other side of firewall from PCM, and way below what Chrysler calls “cabin node” and its Toyota’s equivalent. However, it should be noted that in Chrysler’s case a Bluetooth module is below the inlet of the tube, because it’s mounted below the glovebox, so it could be damaged. Chrysler also lets water exit on top of the transmission. Not sure if that is an excellent idea.

  • avatar

    Wow, my 1987 Isuzu Impulse Turbo had that same problem.
    Impressive to see it again 25 years later.

    And a friend’s Porsche Cayenne flooded several times, apparently due to drain hole issues, which apparently are not listed as a maintenance procedure in the owner’s manual, according to him.

    Why sweat the small stuff, eh VW?

  • avatar

    Honestly as much as I understand the concern of a problem now being replicated across multiple brands or model lines, as long as VW doesn’t completely screw the pooch none of it matters much. I can’t speak for Europe, but VW and Audi buyers in the US won’t let multiple small things stop them from buying. Heck the VW buyers never let big problems keep them from continuing to buy.

  • avatar

    Sounds like improper HVAC drain tube maintenance on the part of no good lazy owners to me.

  • avatar

    It’s a VW. Modern VWs ARE a continuous cascading failure. Wet feet? Cold feet before purchase would have been better.

  • avatar

    Its amazing how these guys have been making cars and still get basic stuff like condensate lines wrong


    • 0 avatar

      My Silverado was recalled because the transmission overflow was routed above the exhaust pipe which could cause a fire. GM has been building cars a lot longer than VW and they can’t get it right either. I’m sure every manufacture has problems like this.

  • avatar

    GM cars had this problem a lot. As far as I know, it was endemic in L, J, and N-body cars, a fault with assembly robots burning holes in panels to let water in?

    Dunno if it’s that on my car or rotted out cowl seals letting water in at the base of the windshield.

  • avatar

    A common architecture or platform does not increase the CHANCE of a failure occurring. In fact, it may reduce it as there are fewer separate different parts and configurations, and presumably the ones that are used can be more thoroughly evaluated and tested.

    What it does do is increase the IMPACT of a failure occurring, as there will be more individual vehicles with the same part and configuration.

    It’s not quite the same thing …

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • FreedMike: “Why would anyone trash a truck for being poor at performance driving when that’s not why they are...
  • redapple: V 6- get them while you can. Diesels,,,? Love them but I m not sure if the DEF, EGR, catalyst choked down...
  • EBFlex: “ Airstream and their parent are testing powered trailers with their own batteries and motors that could help...
  • EBFlex: No. Just accept you are completely wrong. Right now ford loses money on every fake mustang that is sold. What...
  • Astigmatism: It’s all part of the same evil conspiracy that leads some people to buy Fords when you think they...

New Car Research

Get a Free Dealer Quote

Who We Are

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