[Robert.Mauro]

EDIT/UPDATE:
The NHTSA has removed the recall notice and lists ZERO recalls for the Ascent. You may be wondering why. Well, the reason is because all nine customer cars, and all 284 other cars have been cleared and all of them have proper B-pillar welds.

In other words, this is a non-issue and non-item. There are zero affected cars, all cars have been inspected, and there are zero cars being recalled.

https://www.nhtsa.gov/vehicle/2019/SUBARU/ASCENT/SUV/AWD


──────────────────────────────────────

Hey all, I just wanted to summarize (and yes, editorialize) the Subaru Weld Recall for the Ascent. There is a LOT of misinformation floating around out there.

Here's the reality:

• 293 cars had welds that Subaru did not think were up to their ridiculously high standards (I'm forever happy that their standards are so high, which is why I bought the Ascent sight unseen). The welds in question were in a "B-Pillar" (the pillar between the first and second row, between the doors).
  .
• 293 Ascents amount to less than one percent of the Ascents built so far. This happened over the period of six days or less.
  .
• The welds were performed by a misprogrammed robot and caught during their detailed human inspection process. The robot was obviously reprogrammed since then.
  .
• They know exactly which cars from having pulled the robot's logs.
  .
• No one would probably have ever known or noticed (more on this below), regardless, they voluntarily already stopped sale on ALL of the 293 affected cars, and notified the very few customers affected.
  .
• They voluntarily did the recall and stop sale within a few hours of the inspector catching the issue.
  .
• MOST of the cars never made it to customers, and were stopped in transit or on the dealer lots.
  .
• They notified the NHTSA and they (Subaru) officially already filed their own recall notice, on their own, with the NHTSA.
  .
• Any of the very few owners who managed to get an affected car (very small handful of the 293 vehicles) will be given a brand new car.
  .
• The affected cars will be *DESTROYED* - Subaru will NOT attempt to fix the welds.

Here's more info in the official NHTSA filing...

>> https://static.nhtsa.gov/odi/rcl/2018/RCLRPT-18V508-8422.PDF


Here's my Editorializing...
But seriously, think about that for a sec. During the first year, they will have made 60,000 Ascents. This would amount to 4/10 of a percent of them. Only a tiny portion of those ever made it to buyers' hands. I'd say it's reasonable to say that only 1/10th of a percent ever made it to a purchaser? If that many? What's that? 75 people? Less? What's the chances that someone in that tiny pool of people (75 out of 60,000) would be in a sufficiently catastrophic roll-over event that would cause enough stress on the car to make the b-pillar fail? If you really think about it, it's 75 affected cars out of 2.6 million (so you can then look at rollover stats). Then, stop to think about how difficult it is to roll a Subaru compared to anything in its class due to the lower center of gravity, vehicle design and vehicle dynamics system, including our ALL wheel (2 at a time) active torque vectoring system on our Ascents.

I'm not allowed to badmouth other brands (yep, Subaru Ambassadors are held to a higher standard), so, I'll leave this commentary at this: There's case after case of other car manufacturers ignoring FAR larger issues that affect FAR larger number of cars *and people*. There's gas pedals, ignition switches, braking systems... all sorts of catastrophic issues that have caused a number of incidents across other brands where they knew about the problem and played the numbers game. I'm sure you've all heard the stories, and read about the court cases. So, I will leave it at that and not name names or call it anything other than a numbers game.

Subaru, on the other hand, proactively stopped sales, recalled the cars, and are destroying them, to be replaced with new ones. Instead of playing the numbers game, even with the most infinitesimal chance of an issue, they stepped up and did the right thing.

THAT is why I love the brand! THAT is why I felt safe in buying my Ascent before IIHS crash tests the car line. Subaru said they worked hard to make the safest 7-8 seater ever, and, it's because of events like this that I believe and *trust* them.

 

[pilot1226]

Amazing commitment to safety.

Proud to represent the brand.

 

[tariktik]

@Robert.Mauro, you seem to be a good Subie Ambassador. I salute you for that.

Going back to the Weld issue in which my car got recalled for a few days for inspection (I am hoping that the welds are indeed complete and "not missing a few here and there"), would you be able to represent me to your SoA contacts and request that they send me a Dyson as good will? :grin::grin::grin::grin::grin:

 

[traildogck]

@Robert.Mauro

Yeah, your work on this forum (this issue, included) is really incredible.

The amount of vehicles caught is really small, it's it very well documented. And voluntarily transparent. This is the modern world. Back to Robert, Thanks for the work delivering information. Believe me, I know what behind the scenes work means. This community is built on the strength of it's members. Like all 6-Star forums are...

Thanks for the effort and the passion.

@Carl Abrams and @Jason need the same effort metions. Thanks guys for your work from the dealer side.

@FoRealz just keep doing what do.

 

[lazydaze$$]

Hey all, I just wanted to summarize (and yes, editorialize) the Subaru Weld Recall for the Ascent. There is a LOT of misinformation and sensationalism out there, as well as misleading to downright ridiculously incorrect headlines and even stories.

Here's the reality:

• 293 cars had welds that Subaru did not think were up to their ridiculously high standards (I'm forever happy that their standards are so high, which is why I bought the Ascent sight unseen). The welds in question were in a "B-Pillar" (the pillar between the first and second row, between the doors).
  .
• 293 Ascents amount to less than one percent of the Ascents built so far. This happened over the period of six days or less.
  .
• The welds were performed by a misprogrammed robot and caught during their detailed human inspection process. The robot was obviously reprogrammed since then.
  .
• They know exactly which cars from having pulled the robot's logs.
  .
• No one would probably have ever known or noticed (more on this below), regardless, they voluntarily already stopped sale on ALL of the 293 affected cars, and notified the very few customers affected.
  .
• They voluntarily did the recall and stop sale within a few hours of the inspector catching the issue.
  .
• MOST of the cars never made it to customers, and were stopped in transit or on the dealer lots.
  .
• They notified the NHTSA and they (Subaru) officially already filed their own recall notice, on their own, with the NHTSA.
  .
• Any of the very few owners who managed to get an affected car (very small handful of the 293 vehicles) will be given a brand new car.
  .
• The affected cars will be *DESTROYED* - Subaru will NOT attempt to fix the welds.

Here's more info in the official NHTSA filing...

>> https://static.nhtsa.gov/odi/rcl/2018/RCLRPT-18V508-8422.PDF




Subaru, on the other hand, proactively stopped sales, recalled the cars, and are destroying them, to be replaced with new ones. Instead of playing the numbers game, even with the most infinitesimal chance of an issue, they stepped up and did the right thing.

THAT is why I love the brand! THAT is why I felt safe in buying my Ascent before IIHS crash tests the car line. Subaru said they worked hard to make the safest 7-8 seater ever, and, it's because of events like this that I believe and *trust* them.

Admirable reaction to a serious production issue, BUT i have some reservations on an assembly process that would allow this to go undetected for approximately 6 days... company officials must be demanding a precise accounting for the events leading up to the robot programming error and fail safe systems that should have sounded a warning a lot sooner...company CEO must be furious...cost to the company and shareholders perhaps something conservatively speaking...and just a ball park figure...6 million $$...want to bet some “heads roll” on what appears to have been “human error” eventually caught by “human being”. The ambient temperature in Lafayette got a lot hotter! ��

 

[jacket4256]

Thanks Robert for this thread maybe we can get rid of all the other threads ! :grin:

 

[lafos]

Thanks for the clarification on the 293 vehicles. That was the number my dealer told me was related to the hitch recall discussed on this site. Do you think the two are related or conflated? Seems to be some confusion on the hitch.

 

[Burutoo]

Thanks @robert for this detailed info of the recall.
When i read this part:
"The welds were performed by a misprogrammed robot and caught during their detailed human inspection process. The robot was obviously reprogrammed since then. "

I thought you were going to say 'the robot was immediately shamed and then disassembled for scrap'

 

[Robert.Mauro]

lol, scrapped the robot

 

[Carl Abrams]

i have some reservations on an assembly process that would allow this to go undetected for approximately 6 days...

Second try at this - I had a long reply last night and the forum said I couldn't post here then. :tango_face_surprise

Anyway ... if they're doing things like they did when I was there, it takes more than just a couple of minutes to weld up a B-pillar. More like 10. So what they do is sub-assemble them off line, then those that are subbed up are placed on a rack and taken to the warehouse for later use online. Hoods are done this, for example, due to all the complicated welds that need to be made on them.

Now, keep in mind one other detail. We've NEVER had this issue before. We've had our share of problems, obviously, but in all of the production vehicles we've run - and we're talking about hundreds of thousands of them - again, we've NEVER had a B-pillar weld issue before. So the associate working the line would have had no idea that this was even possible. I've seen issues in the past where the welder didn't have enough current or the weld tip was bad, but those result in welds not sticking and are typically caught immediately or by the body shop final inspection line, before they ever get to paint.

Oh, and the other thing is that there are 293 that have to be inspected. That doesn't mean that all 293 of them are bad, just that's the number that could POTENTIALLY be bad. It may only turn out to be half a dozen cars total. That's something we probably won't know for quite a while ... I'll probably know on the three I have waiting at railhead when they either remove them from our pipeline or clear them for sale once the inspection is done, but I have no hope that it's going to be a quick process. Subaru would rather err on the side of caution than compromise their standards.

(Editorialize: I'm not an Ambassador, so I can freely bash FCA and the MILLIONS of vehicles they've shipped out that needed recalls - typically after they kill several people.)

 

[Robert.Mauro]

One update... there's a count being shared about how many delivered cars were *possibly* affected. Of the 293 cars that might have experienced the bad welds (as Carl Abrams said, not all did), there were only NINE that were delivered.

 

[scooby]

One update... there's a count being shared about how many delivered cars were *possibly* affected. Of the 293 cars that might have experienced the bad welds (as Carl Abrams said, not all did), there were only NINE that were delivered.

Great info! Nice stats to have to assure folk that everything is going to be ok.



So very impressed with Subaru and how they're handling this. The fact that only 293 cars were affected and 9 were delivered is pretty impressive. They caught it quickly!

 

[lazydaze$$]

Carl, good to get your comments...If these b-pillar welds are done robotically, computer programmed by somebody, and they discovered a programming error albeit after an inspector detected lack of adequate welds...wouldn’t computerized production accounting records almost certainly be confirming the lack of welds in all 293 vehicles? IMHO. I think they have almost certainty regarding the magnitude of the problem. If they don’t, then that’s another issue that needs addressing.

 

[norway]

Carl, good to get your comments...If these b-pillar welds are done robotically, computer programmed by somebody, and they discovered a programming error albeit after an inspector detected lack of adequate welds...wouldn’t computerized production accounting records almost certainly be confirming the lack of welds in all 293 vehicles? IMHO. I think they have almost certainty regarding the magnitude of the problem. If they don’t, then that’s another issue that needs addressing.

There are so many variables that speculating without knowing all the details, and essentially being the people involved, is a bit of a big guessing game. For example, what if the issue with the robot meant it didn't fail all, as in it was a tolerance issue where it was potentially a cold weld joint on some, or an alignment issue where it could go from there to partial to missing, etc. Or it could simply be it skipped/missed them. Regardless they will want to inspect all that could possibly have been affected to verify, only way to be sure.

 

[Carl Abrams]

One update... there's a count being shared about how many delivered cars were *possibly* affected. Of the 293 cars that might have experienced the bad welds (as Carl Abrams said, not all did), there were only NINE that were delivered.

Here's an additional update.

I had three cars sitting at railhead that were part of the 293 and were on the recall yesterday.

Those three cars today now show cleared and ready to ship to me. So it really is a case of apparently affected versus actually affected, and Subaru being incredibly cautious. That's not helping with the slowness we're seeing in getting these cars actually completed, but what I see happening is that Subaru plans for this car to be in our fleet for a LONG time and they're getting all the bugs out of the line now. I mean, I remember how much fun it was back when we first started building the Legacy and Legacy wagon, and all the inspections we had to do then. Seemed like every weekend we were going through the lot outside and doing inspection campaigns for clips, missing torque paint, or whatever. And that was when the factory was brand new.

 

[Freddo80]

Thanks for the information!! Any lists of the potentially affected vehicle VIN #? It would be nice to check up on my order and see why it is taking a month longer to receive. Notified dealer and doesn't know squat.

 

[tariktik]

One update... there's a count being shared about how many delivered cars were *possibly* affected. Of the 293 cars that might have experienced the bad welds (as Carl Abrams said, not all did), there were only NINE that were delivered.

I guess I was one of those "lucky nines" whose car got delivered (albeit passed inspection). It's like winning in a lottery. LOL. SoA really needs to send me something. :grin::grin::grin::grin::grin:

 

[asbaker]

Concerning your saying "buying my Ascent before IIHS crash tests the car line"

How long does it take for the crash test to be done and published? On my 2014 Forester (bought in April 2013 but called 2014 due to some changes) the insurance company charged a higher rate because that had been changed from 2016 model and did not have the test results when we bought it. I don't know if our insurance will change our rate after the test. We picked our Ascent up July 9th.

Thanks Robert for all your clear information. You answered me before I even asked my last question concerning the bolts in earlier delivered cars like mine.

 

[Tishers]

Robotic welding heads monitor the arc current on every contact point. That is part of how a robotic welder knows that the welder contacts are good. Often those welding machines have a caddy with spare heads or contacts that they can automatically swap out when one is being worn out or fails.

Where the problem comes in is with the human part of the progamming of the PLC (programmable logic controller). As the head will position itself several times for different welds the engineer can set the welder to trigger only on some of the weld points and not "all guns at once".

If a human does not program in the equivalent ladder logic instruction of "weld 1 3 4 5 7 9" and drops or omits the code to say "weld 1 3"... Then welds 4 5 7 9 don't get made and the machine cannot recognize that it was a mistake.

There are literally hundreds of thousands of lines of ladder or STL (statement list) code that is being maintained by 30-40 controls engineers... In this case, a mistake was made and it did not get caught.

( controls engineer with 35 years of experience in process automation)

 

[Subiesailor]

Way back in the day my HS Auto Shop received two brand new Toyotas. They were part of a similar welding flaw recall. They had to sign all sorts of paperwork stating the cars had to be destroyed after x number of yrs. Never driven off the campus etc. It was a really big deal given it enabled the program to teach modern vehicle design. Maybe Subaru can do something like that vs scrapping them?