Toyota Hoped The bZ4X EV Would Usher The Brand Into The Electric Era. Then The Wheels Fell Off

toyota bz4x Recall Topshot

It’s no secret that Toyota has a pretty big problem with its new bZ4X electric crossover, the company’s first high-volume electric car in the U.S. With no fix for the EV’s wheel bolt woes in sight, Toyota’s now offering buyback deals to American bZ4X owners. Toyota has sent owners a letter regarding these buybacks, and here’s what is has to say.

…Toyota will offer to repurchase your vehicle. The terms for the repurchase may vary, depending on your state and particular circumstances.

In addition, the compensation package for bZ4X owners who wish to keep their vehicles has grown to be quite vast. We’re talking about a free loaner vehicle, reimbursement for fuel for said loaner vehicle, a $5,000 credit on the vehicle, complimentary charging at EVgo stations through 2024, and a warranty extension.

More details on compensation can be found in the above letter, confirmed by Engadget to be legitimate. As for American owners of the bZ4X’s Subaru Solterra twin, they might not really exist yet. The initial NHTSA recall report states that “Subaru has informed Toyota that no Solterra vehicles have been sold to dealers or retail purchasers.” So how did we get to the point where Toyota needed to send a buyback offer to bZ4X owners? Well, it’s a little bit complicated.

Solterra Wheel Bolts
Photo credit: Subaru

First, let’s define the issue Toyota’s having with the bZ4X. While “hub bolts” is the terminology most associated with this issue, it’s worth clarifying that the recall has to do with wheel bolts like you’d find on many European cars. Should these bolts back off, the wheel would detach from the hub, rather than the hub detaching from the knuckle. Subaru issued a lovely diagram detailing the issue, which I’ve inserted above. It’s very strange language in an age of bolt-on hub assemblies, but that’s the way it goes sometimes. So how did Toyota find out about this issue? Let’s take a look at the defect information report uploaded to the NHTSA website.

Mid-May 2022 – Late June 2022

Toyota received a field technical report from the Taiwan market indicating separation of the front left wheel from the vehicle. Toyota recovered the wheels and hub bolts from this vehicle for investigation.

Toyota then received two field technical reports from the U.S. market. One of the reports indicated loose hub bolts of the front left wheel and the other indicated a separation of the front left wheel from the vehicle. Toyota reviewed the production process at the vehicle assembly plant and did not identify any abnormalities on the hub bolt tightening history of these three vehicles. Toyota also reviewed the assembly process of the wheel and hub bolts at the assembly plant and did not identify any abnormalities. In addition, although damage was observed on the wheel recovered from the vehicle in the Taiwan market, the damage was believed to be a result of the wheel separating from the vehicle and the cause of the loosening of the hub bolts was not able to be identified. Toyota continued to investigate and conducted a driving test using mass production wheels and hub bolts that were tightened to specification. The testing showed that the hub bolts loosened under certain severe driving patterns.

Wow, that’s not good. Toyota recalled the bZ4X in late June, which means that we’re more than a month in without any updates on a potential fix. At the time, owners were advised to avoid driving their new electric vehicles, so all affected models should still be grounded for now. It’s more than a little bit weird that Toyota didn’t discover this issue during development. From climate testing to final calibration, development engineers often puts tons of miles on prototypes. Given how early we are into the production run of the bZ4X and its Subaru Solterra twin and how these failures have occurred on low-mileage examples, shouldn’t self-loosening wheel bolts have been caught and corrected?

Wheel Bolts

Weirder still, this isn’t Toyota’s first attempt at using wheel bolts instead of studs and nuts. For instance, current Lexus IS models in non-V8 trim and the Lexus NX 350h use wheel bolts with a part number of 90942-02090. Plus, other automakers have used wheel bolts on EVs without issue. The BMW i3 and Volkswagen ID.4 immediately come to mind. There’s no real reason why the use of wheel bolts might create more issues than a stud and nut arrangement, so the nature of the defect is quite puzzling. As it sits, Toyota doesn’t have a fix for the bZ4X despite the initial defect report having been issued more than a month ago. So what gives? Well, a fix could require new hardware and additional testing, which may push back the timeline for a solution. Plus, if any component changes are required, today’s tight supply chain might affect Toyota’s ability to procure new components.

2023 Bz4x Limited Awd Heavymetal 042
Photo credit: Toyota

In any case, this recall is a bit of a disaster for Toyota. While the Japanese brand has dabbled in compliance EVs like the Scion iQ EV and both generations of the RAV4 EV, Toyota’s generally dragged its feet on producing a mainstream battery electric vehicle. With an incoming EV offensive, the bZ4X was supposed to be the flag-bearer, but its failure to launch has tainted Toyota’s reputation in EV circles. While a bad reputation in EV circles may not seem like a huge deal to some, it’s worth noting that EVs are often enthusiast vehicles. They may not pack the tire-shredding boisterousness of muscle cars or the poise of mid-engined sports cars, but EV owners are passionate, informed, and often the first point of peer-to-peer EV advice for many potential buyers. This bZ4X wheel bolt recall is on par with Ford Focus RS head gasket issues and likely won’t be good for Toyota in the long-run. We’ll continue to watch the bZ4X recall closely. Hopefully owners who wish to keep their Toyota EVs can find some form of resolution soon.

Lead photo credit: Toyota

Share on facebook
Facebook
Share on whatsapp
WhatsApp
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

74 Responses

  1. My 1973 Mazda had wheel bolts.
    It had locating pins so I didn’t have to hold the wheel up to thread the bolts.
    On the rear, those save pins threaded into different holes to push off the drum brake.

  2. I had a BWM 2 series that was an absolute blast to drive. One day when I was getting ready for a long and spirited road trip I was checking my tire pressure and noticed that one of my wheel bolts had gone missing. Since then I have made it a point to retorque my wheel bolts every time I check the air pressure (every other month or so) which in the last 4 years has only turned up one loose nut, but between that one and the one that went awol I find it prudent to continue checking.

    1. European here – Mercedes does bolts (My w205 had ones, and it was a PITA to find correct ones for aftermarket wheels), as does some BMWs. Pretty strange for Toyota to go that way, since all (I guess?) Japanese manufacturers use studs and nuts (Honda, Mitsu and Suzuki for sure since I had cars from them).

      1. I HATED changing the wheels on my VW because they used bolts instead of studs and nuts. Of course, I didn’t hate it enough to pay somebody else to do it so there is that.
        I wonder why the Germans stick with bolts.

    1. My Saturn Astra had Wheels Bolt too. (since it was 99% an Opel).

      Airstream recently had a recall due to axle washers in axles provided by Dexter. The washers would break and the wheel would come off. They caught it quick and it only affected 250 trailers (many of which hadn’t been delivered yet), but a few people did lose wheels prior to the recall. Sometimes it’s the simplest thing that has been used for years, but something minor changes, or you get a bad run of parts and it becomes dangerous. A wheel coming off a vehicle is a pretty dangerous situation.

      1. I’ve had the rear wheel come off a motorcycle at speed, actually the chain wrapping itself the engine was the exciting part but still. I’ve had the front wheel fall off of a Volkswagen, In that case the driver side CV joint seizing up and tearing the Macpherson strut in half before it threw the car into the oncoming traffic was the exciting part. More than loosing a wheel per se, but still. A couple years later I had both rear wheels fall off a car simultaneously because the idiots at the tire shop torqued the nuts on the mag wheels while the car was resting on the ground. When the one wheel came off that was enough to pull the other off.
        Still not as bad as loosing your brakes or catching fire.

        1. My dad had a wheel fall off his Chevrolet Corsica at about 30mph (tire shop, of course). He saw it roll ahead of him, managed to gently turn rightwards while braking, and stopped with no other damage whatsoever.

    2. Its literally a bolt. The head normally has a bit of a taper to fit in the recesses on the wheels, but its just a bolt.
      A lug stud has a threaded portion sticking out, the lug nut goes on the stud to hold the wheel on.

      Here’s a BMW with a similar setup to what’s being discussed here: https://f87.bimmerpost.com/forums/showthread.php?t=1683573

      As for why they did it, there are a lot of possible reasons. Hanging a wheel on a stud and pushing it on is a good way to damage the threads or the wheel. It may give a mechanical advantage to fastening. It is theoretically easier to install different wheel sets onto the same suspension package.

      Several companies that used to use lug bolts in the past have gone back to lug studs. It just depends on what’s going on internally

    3. My Sportwagen has them. I figured it’s just because the Germans have to make everything fiddly and stupid to remind everyone what great engineers they are.

      I can’t comment on any benefits, but as far as negatives, you will either need three hands to put on a wheel so you can hold it with 2 and start a bolt with the third, or you can buy a metal rod that threads into the lug bolt hole to hold the wheel. I use 2 to improve centering. You also can’t reasonably access the back of the hub, so any corrosion or other crud on the tips of the bolts is getting dragged through the threads.

      I expected more intelligence from Toyota.

    4. I’d like to know too. From a usability POV having a stud to hang the wheel off of when installing seems better. But other than that I don’t understand why one system would be better/worse than the other at keeping the wheels attached.

    5. Old VW bugs had studs instead of nuts, sometimes you had to smack them with a hammer to loosen them up. And some old Chrysler products had reverse threads on the drivers side of the car. I dont know if it was better, but reading this article, I would think not.

        1. This is of course because fretting-induced precession (wiggly spinny motions) will cause normal bolts or nuts to want to unscrew themselves on the passenger side of the car. Interestingly, if you make the interface between the nut/bolt and the wheel conical instead of flat, the phenomenon disappears. This is why lug nuts are normally tapered, and fit into a matching recess.

        2. Saabs, having been designed by aerospace engineers, did the same thing in various locations for a long time but finally realized it was causing more problems than it could ever solve. Some wheel bearings, a screw for the water pump in their B-series engines, and other things. Bentley and Rolls-Royce did the same.
          Studebaker used reverse threaded studs on the left through 1960-1961.

      1. Because some engineer prefers them and he/she had enough clout with the relevant design team to make it happen. Either that, or it saves 2.7¢ per fastener because instead of having to make a threaded hole and then put a stud in it, you can just make a threaded hole.

        1. Wheel studs are pressed into hubs from the back, so normally hubs aren’t threaded.

          If you use wheel bolts you do actually have to thread the holes in the hub so it is probably more expensive.

  3. I know it would cost perhaps an extra dollar a car, perhaps $10 for a retrofit, but wouldn’t safety wiring be a relatively cheap and easy fix? I mean, compared to a buy back, drilling and safety wiring the bolts is cheap, easy, and well safe.

  4. This is an amazing problem. I’ll bet it has to do with the torque profile of the electric motors combined with precession along with design of the bolt taper. Basically, if you try to fully enjoy the instant torque hit from the electric motor, it induces just enough flex and/or play in the tolerances to put an infinitesimal twist in the bolts which then gets released by the threads. See what I’m saying here? Think of an impact driver. I’ll also bet a throttle remap is part of the eventual fix. I also wonder if it happens to the right front tire on right-hand drive cars due to the weight of the driver offering more traction to said tire. Anyway, I’m wrong a lot, but that’s my guess.

  5. Sometime in the last century I helped a lady change a flat tire on a Beetle in the dark. I had never seen wheel bolts before I hoped I never would again. I managed to keep all the curse words in my head. To this day they leak out every so often. If Toyota is willing to buy back the cars, good for them. BTW, am I the only one who thinks, “Busy Four Times?”

  6. My VW T4 Transporter has wheel bolts… they’ve never been an issue over the past twenty years. The only trick is that there’s two different part numbers – one for steel wheels and one for alloy wheels.

    As for Toyota, they have form on this with 70 Series Land Cruisers. Especially in their transition to the V8 turbo diesels. Turns out the wheel nuts were too thin and, if the vehicles were operated close to GVM in rough environments (like mine sites or long distance touring), they could come loose. They then offered thicker mine spec wheel nuts with plastic indicators… problem went away. Took many nuts…

  7. I had the front left come off my r53 Mini because the guy I had just bought it from got the wrong length bolts. They were a half inch short so only one or two threads were actually holding on. Aparently such a small amount of thread didn’t like having two fat Hawaiians (my brother and myself) tossing it around the couple roundabouts in the city I lived in at the time. Replacing the shitty aftermarket bolts with factory ones solved the issue on the wheel, but when it dropped it fell catching the wheel with the hood so that got all bent to hell.

  8. [Toyota Spokesperson:] It’s a great pleasure, thank you.

    [Interviewer:] This car that was involved in the incident on the highway this week…

    [Toyota Spokesperson:] Yeah, the one the wheels fell off?

    [Interviewer:] Yeah

    [Toyota Spokesperson:] That’s not very typical, I’d like to make that point.

    [Interviewer:] Well, how is it untypical?

    [Toyota Spokesperson:] Well, there are a lot of these cars going around the world all the time, and very seldom does anything like this happen … I just don’t want people thinking that our cars aren’t safe.

    [Interviewer:] Was this car safe?

    [Toyota Spokesperson:] Well I was thinking more about the other ones…

    [Interviewer:] The ones that are safe,,,

    [Toyota Spokesperson:] Yeah… the ones the wheels doesn’t fall off.

  9. I have an R55 MINI, I rotated my tires one day and a few days later heard a weird noise in the left front…..the bolts (Yes, BMW influence) were loose. I knew I had torqued them, but I just shrugged and re-torqued them again. A few days later – damn, bolts are loose again! But only on the left front, all the rest were fine. What the heck was going on?

    So the answer was simple – when you rotate tires you usually just move them front to back and vice versa, I had gotten too smart for my britches and cross rotated the front tires. So fronts went straight back, and the rears crossed as I moved them to the front. Now normally this would not be a problem but in this case I had not noticed that the tires I was running were directional. I crossed the fronts back over and all’s good – lesson learned!

    I’m guessing that the directional tread set up some sort of sympathetic vibration that caused the bolts to loosen when they were turning the wrong way?

  10. OK I may be off the mark here because I don’t know what these bolts look like, but it reminds me of an issue we had at work on some manufacturing jigs.

    We used some heavy duty zinc-plated 12.9 class bolts. We torqued them to spec, left the assembly on a crane overnight and when we came back the next morning half of them were loose. Imagine the face of the dude who saw 8 tons of steel hanged on a crane with half its parts coming off!

    The torquing procedure was fine, but it turned out the plating induced hydrogen embrittlement in the bolt, causing them to relax when a sufficient load was applied a certain amount of time.

    Now we recheck every bolt 24h after it’s been torqued.

    1. If the bolts had hydrogen embrittlement then the plating process was incorrect. They need to be heat treated after plating to address the hydrogen embrittlement. Also all hydrogen embrittlement I have seen causes cracking and failure, so would not be corrected by a second torque. I have seen far more issues with washers failing for this reason. We had an issue some years back where we would torque the bolts and within a couple of hours you could walk past the unit and hear the sound of washer pieces hitting the floor. Obviously at that point there was zero clamping force.

      1. Yes the plating process was definitely at fault, but the “lesson learned” that was written back then definitely referred to the embrittlement of the 12.9 bolts.

        I wasn’t employed there back when this happened, but I believe there has to be a link between the plating and the bolt class as you won’t find a French bolt manufacturer who’ll sell you plated 12.9 bolts while 10.9 and lower are available plated.

        I can’t tell you how the bolts looked like after this event, I’ve only read about torque issue, not damage but again: it’s not first hand knowledge.

        1. Yes, 12.9 are more susceptible to hydrogen embrittlement as they are harder to start with. The SAE equivalent would be L9 which has the same issue. Most of the standards I am familiar with recommend that bolts over 1″ not be plated at all due to concern for the potential problems.
          I have a few bolts that failed like this on my desk.

    1. My guess is that they’re going overkill because they’re worried about the reputation damage from their rather late entry to the electric market being seen as unreliable, when the perception of reliability was basically the only selling point it has over other options.

    2. There’s a few possibilities. One is that something in the design is so off that nothing short of a complete redesign will keep the wheels on. Those take time, and producing new parts is logistically difficult.

      Another possibility is that the specific driving condition isn’t an expected behavior of the car, and they’re trying to determine a way to eliminate the behavior, which also takes time.

      And finally, lawyers are involved.

      1. If you look at the Loctite Adhesive Sourcebook, they have different products for various hardware sizes. For example, 271 is high strength for bolts up to 1″. 277 is permanent for bolts over 7/8″. They do have some products that are only recommended for smaller hardware.

  11. In my country, there a superstition where if a child easily get sick and there’s no medical treatment that could provide a long term fix, people believe that the child bears a bad name. And believe it or not, changing their name work wonders. I’m not a superstitious person, but I’ve witnessed such cases too many times (including 2 of my cousins) to not believe it.
    So yeah, Toyota might want to consider it.

      1. We use Loctite on hardware all the way up to 1.5″. I think the reason they would not use Loctite would have more to do with it being fine in a manufacturing setup, but how would you ensure that a customer would use Loctite each time they swapped wheels on the vehicle?

        1. Huh. We start using lub for M12 and up. M10 is a bit of a gray area. I believed it’s better for the clamping force past a certain point to ensure good lubrication through anti seize than relying on Loctite to keep things in place.

          Maybe that’s just something we do internally and not an industry standard.

          1. Using lubricant for torquing is a very good practice as it reduces the variations in friction and allows for more consistent clamp load values. You just need to be aware of the lubricant K factor and reduce the torque applied by the correct amount. At one point we used a EP2 moly grease left over from barrels in grease machines to protect wheel studs from rusting during shipping. Some customers didn’t follow our instructions to clean the grease off prior to torquing the wheel nuts and maintained the dry torque value. Lots of wheel studs ended up failing to the point where we had to stop using grease for rust protection.

            1. Yes that’s something I’m aware of. Facom produces a good little cheat sheet for torque specs, including how to account for bolt class and lubrication.

              I don’t know precisely of the K factor you talk about, but I presume it relates to the friction losses when torquing the fastener that do not translate to clamping force.

              I wanted to try and model analytically how torque translates to clamping force accounting for these parameters and bolt geometry but it was more time consuming than anticipated and my work load got in the way.

              If you’ve got some reference material related I’d be interested

        1. It’s says they aim to inform & entertain. I, personally, feel they do both. Given the often facepalm-inducing—f not downright heart wrenching—news these days, I’m happy to stumble across a few groaners to help lighten the mood.

          Your Mileage May Vary—wildly

Leave a Reply