The Tesla Cybertruck has generated a great deal of interest thanks to its oddball design, both inside and out. It’s not just a pointy weird triangle, it’s also one that relies on unique materials and engineering solutions. Finally, we’re starting to see teardowns reveal more of the truck’s secrets that are hidden under its skin.
YouTube channel Autoline Network headed down for an interview with Caresoft president, Terry Woychowski. If you haven’t heard of Caresoft, they’re a company that specializes in automotive benchmarking and cost consulting.
What does that mean? It means they’re experts in teardowns and analyzing vehicles from an engineering perspective. Right now, everyone’s interested in the Cybertruck, and they’re in the middle of tearing one down. There’s already plenty to see inside the body of the beast.
As the video shows, Caresoft has two Cybertrucks on hoists right now. One’s still intact and is being used for a variety of testing purposes, but the other has been pulled apart to a great degree. With the doors off and the battery out, one shocking thing stands out. There’s no floor!
As it turns out, the battery of the Cybertruck is the floor. “With the integration activities that were really kind of a paradigm that was broken by Tesla on the Model Y, is that the body-in-white has no floor,” says Woychowski. “It simply takes the top of the battery, and then when they install the battery pack it becomes the floor of the vehicle.” All the carpets and trim sit on top of the battery inside the vehicle.
In fact, the battery sits in the Caresoft workshop, it still has the front seat mounts sitting right on it. They’re cast parts, according to Woychowski. The rear seat support is a casting too, albeit attached to the bulkhead, not the battery.
Interestingly, most of the cast parts on the vehicle, including the large gigacast sections, are painted with some kind of vaguely brown-yellow coating. Woychowski speculates that this is for corrosion protection reasons. This is notable as cast aluminum parts don’t usually require any coating. Aluminum forms a tough oxide layer that protects it from further degradation. Many cars have aluminum oil pans or transmission housings that have no coating whatsoever. For some reason, though, Tesla considers it important to coat these parts. It’s also worth noting gigacast parts in the front end appear to have a black coating for aesthetic reasons.
Another key focus for the Caresoft team is learning about the 48V electrical system used in the Cybertruck. Making that job easier is the fact that wiring harnesses are clearly marked in the vehicle. High-voltage lines are bright orange, as is typical in the EV world. Meanwhile, harnesses for the 48-volt system have blue connectors and blue wraps at various points. Notably, though, Woychowski notes that there are still some traditional 12-volt circuits on the car, without being specific as to which.
“We have to go deep into every electrical system, because it’s not all 48 volts,” he says. “Some of it remained at 12 volts.” The team’s goal is to inventory the truck to identify exactly which systems use 12 volts and how that voltage is converted down within the vehicle.
Steer-by-wire is also a notable inclusion on the Cybertruck. There is no steering shaft connecting the steering wheel to the steering rack. The teardown shows the sizable motor under the dash that’s used to provide feedback to the driver through the steering wheel. This feedback is important for feel. Without it, the driver can’t intuitively understand which way the wheels are pointing or what they’re doing. Typical steering feedback includes a degree of self-centering torque to bring the wheels back to pointing straight ahead and a level of resistance to driver input. This mimics what happens with a traditional mechanical steering system.
Meanwhile, on the steering rack itself, there are a pair of motors to translate input from the steering wheel into motion of the front wheels. “You really have to design a system like this to say, it can’t fail, it can’t have a single point failure,” says Woychowski. Hence the two motors, for redundancy. We learned about this previously when Sandy Munro interviewed Tesla’s engineers about the steer-by-wire system back in December. Each motor is sized to provide roughly 50-60% of the peak torque required in low-speed parking situations. If one fails, you’ve still got one more that can handle regular steering duties well enough to get you off the road.
The 48-volt system comes in handy here, as it makes it easier to supply plenty of juice to the feedback motor and the steering motors alike. At the rear, a similar rack is used to steer the back wheels. However, it only has one motor as redundancy is not so important in this case.
With the intact Cybertruck, the Caresoft staff demonstrate the low-speed steering mode where the rear wheels steer opposite to the front wheels. This cuts the vehicle’s turning circle in parking situations and the like. At road speeds, the rear wheels instead turn the same way as the fronts to allow for smooth lane changes and better stability.
Interestingly, Woychowski notes that the Cybertruck initially wouldn’t steer on the hoist. The vehicle was able to detect it was being jacked up, and wouldn’t steer. Caresoft technicians were able to find an unspecified way to put the truck into a special mode such that the steering would operate on the hoist.
There were a few build errors found in the truck, too. Woychowski first points out the driver’s side door seal, which is puckered in one section and not perfectly seated. He also notes that one of the rear doors was incorrectly fitted with a hinge intended for use on the front doors. This prevented it from opening as far as the properly-built rear door assembly on the other side.
Curiously, owners have been reporting door issues:
I "fixed" our door panel gap. I doubt this is a "5 minute" service job but it shouldn't be too long.
Moral of the story, gaffers tape heals all wounds. https://t.co/dz6Am7H6WN pic.twitter.com/IK3n7VFf33
— Miles Somerville (@SomeOfMiles) March 20, 2024
There’s still plenty to learn about the Cybertruck. Cost analysis by Caresoft will tell us whether the 48-volt changeover makes sense, and how many systems are still lagging back on 12 volts. In time, we’ll also likely learn more about the value of gigacast sections and whether or not steer by wire is as reliable as we’d like it to be.
For now though, this is already a great look inside the Cybertruck. It shows us just how different EVs can be from the ICE-powered vehicles of old. How times change.
Image credits: Autoline Network via YouTube Screenshot
Steer by wire with no physical backup? Jesus, I wouldn’t ride as a passenger in one of these things for 5 minutes.
Honda Prelude managed rear steer like that with mechanical components.
Throttle by wire, fine, it’s been done for a long time.
Brake by wire, fine, with redundancy in-place.
Steer by wire……..I think that’s where I draw the line. Mechanical linkages can fail too, but I’d still feel a lot more comfortable with a physical connection. Maybe I’m paranoid, but I’ve been a driver and a passenger in vehicles that had complete electrical shutdowns (gas and EV, with LV and HV shutting down) while going over 60mph.
If you have a 12v/48v/800v issue (that could just be software issue) and all of them shut down… what happens with steer by wire in the Cybertruck going down a highway?
The great thing about using a big steel rod to steer your car is that mice can’t chew through it.
You haven’t seen the mice in my place.
I didn’t even think about that, but also a good point.
Denver Airport has a problem with the huge amounts of wild rabbits that chew through wires in the economy lots. They seem to like Audi/VW products most, not sure why… but there’s gotta be a joke in there somewhere.
Steer by wire. . .Nope. As an Electrical Engineer, drive by wire, how it is implemented, scares the crap out of me.
That said, I have had the steering knuckle break on us in a truck. Lucky we were off roading and doing a multi-point turn when it happened. So, I guess even a single mechanical failure can cause loss of control.
Other thought. . .seems like a lot or work and engineering to make that work? Almost seems like a simple, conventional steering system would be more cost effective. It is lot of components to replace a 3′ piece of steel and a few knuckles.
I hitched a ride in a big rig cab (only) after kayaking and as we were going down the twisty turny road I asked what would happen if the power steering failed. We’d go off the edge was the deadpan response.
Tesla’s implementation may be bad, but steer/fly by wire has been a thing in aerospace for a while, and it’s way, way more reliable than the mechanical linkages and cables it replaced. I used to have the same thought that I would never trust a solely electrical solution, but if you look at the numbers for mechanical controls, they are far worse in every regard.
Of the 26 cars I’ve owned, 23 had throttle cables, 3 had electronic throttle bodies. Guess which system has left me walking? (Hint: It wasn’t the throttle cables.) As badly as the automotive industry has screwed up throttle-by-wire (Hyundai/Kia minvans and SUVs circa 2010 anyone?) repeatedly, I don’t care how reliable aircraft fly-by-wire systems are, the automotive industry is NOT held to the same standard, and steer-by-wire without a mechanical connection as a redundancy is a horrible idea.
Those QA inspection marks on the connectors are odd for a new build. Usually they indicate that a quality issue was identified, and inspection had to be done for all finished products. Multiple marks is not a good sign.
I like the Porsche accelerator pedal
also reminds me of the one on the Holden Hurricane
Some goober’s gonna try to mount a powered sub under the rear floor and drill right into the battery pack, sure as a world.
I renovated a bathroom in my old house. The previous owner did the shower, and when he screwed the backerboard to the studs, he put a screw into the shower supply pipe. Was a very slow leak for 15 years. Enough to rot the floor, but not enough to cause REAL structural damage.
You are totally right.
Horrifying but quite possibly true
Yup. Just like the installer that drilled into a gas tank when I worked at the hifi shop.
It already has 2 powered subs. Not sure there is enough room under the seats for better units.
Please remind me to hold onto the driver’s door when turning right lest it fly open. 😀
So it turns into a crematorium on wheels if the battery/floor ignites?
Yeah some fine engineering here. /s
I doubt a floor will be much help against a battery fire anyway.
Should have made my sarcasm clear. My bad.
Aluminum forms a tough oxide layer that protects it from further degradation. Many cars have aluminum oil pans or transmission housings that have no coating whatsoever.
Cast aluminum parts do indeed corrode. It takes a while but look under the hood of any older car and you’ll see it.
Also the oxide layer is very thin and easy to wear through, especially in high traffic areas like floor pans.
The Delorean had a potential issue with the engine that it would corrode in the valley and through some oil channels completely ruining the engine.
I look forward to watching that video when I have time. Until then, I’m curious about the redundancy designed into the steering – does the redundancy go from end to end in the system? E.g. two motors won’t save you if, say, they get power from the same circuit and something else in that circuit fails in a way that cuts power to the motor(s). I’m also curious whether they ever considered using the front suspension castings at each corner in the rear to save money.
It will take a few years of owner Beta testing (or are they the Alphas?) CTs on the road before many of us would have enough confidence in them to consider owning one. Going blind would also help… but we’re just considering the mechanicals here.
If they are as dependable as Full Self Driving is, then kindly count me out.
May I make a suggestion for the website as a whole? As most of your images are higher resolution than presented on the page, can you make them clickable so they open in a new tab or in a previewer of some sort at native resolution? I know I can and frequently do right click and open in new tab, but there’s probably many people out there that wouldn’t think of doing that.
(Personally I prefer opening them in a new tab over a previewer, but I expect to be the minority)
Amen. I right-click and open in a new tab as well, but making them clickable would be great.
Yes, please do this. It would make the pictures much easier to see at their higher resolution, especially on mobile.
Yes please!
We did some hvac testing on a van back in the early 00’s that was 48v for many systems. It looked like what would later become a transit or sprinter van, it has lots of masking on it even while it was in the climate controlled test cell running. Basically a chassis dyno that we could get get the temps down below zero and up into the 100 deg to test how well the hvac systems performed in different conditions.
“You really have to design a system like this to say, it can’t fail, it can’t have a single point failure,”
SPOF is my main complaint about the ” tabletizing” of far too many critical features on just about every modern vehicle. Airliners have had digital dashes for decades, but are required to have triple redundancy, with an electrically isolated, electronically independent, back up instrumentation for critical functions. Tesla in particular has marketed their OTA updates as value added, when in fact it is an admission that they are releasing to our public roadways systems that have not been fully vetted, and I believe there is a greater than 0 chance that an OTA update could result in lock-up/blackout situation.
So many things are questionable about the CT that it seems it has been released prematurely.
We had a model 3 for a few years. The screen blacked out while my wife was backing up and using the backup camera. Not a good experience!
I was initially a fan of the OTA updates, but there were a few cases where they moved safety critical things like the current gear, current speed, and speed limit to different areas of the screen (or changed how they appeared). It turns out I don’t like change that much, when you mess with the controls on a 4,000 pound vehicle with life-or-death stakes.
I think you need muscle memory to be a safe driver, and Tesla constantly fiddling with controls to cut costs (like the touch screen shifter and moving blinkers to a stalkless design on new vehicles) is really worrisome.
The battery is the floor? Yikes.
Yeah, I think Faith No More first had the idea.
“LET THE BATTERY BE THE FLOOR, LET THE BATTERY BE THE FLOOR”
So it’s put together as carefully as a late ’70s Malibu, but we’re supposed to trust the video-game steering system? Hell no. I’ll never even ride in one. And I’ll give them a wide berth on the road.
Oh come on. Properly sketchy stuff just doesnt make it past the safety nazis at NHTSA.
Please tell me your comment is sarcasm
Nope. I really do believe that the regulators do a very thorough job of catching unsafely poor designs, at least for the things that they check.
What makes you think otherwise?
I was responding to “safety nazis”
What should have been sarcastic about “safety nazis”? I say that because NHTSA safety Nazi action has prevented many non-dangerous things like non-sealed beam headlights, pop up headlights, and adaptive headlights.
To be faiiiir, Tesla at least put more redundancy into their video game truck’s steering than Boeing put into systems designed to keep planes from literally falling out of the sky.
#FACTS
It wasn’t designed to keep them from falling out of the sky, it was designed to make them appear to handle exactly like an NG so pilots wouldn’t have to retrain, the Max is flyable without MCAS at all (but not as much with a malfunctioning MCAS) and there’s been some debate as to whether the system is even needed at all, without MCAS, the Max still handles more similarly to the NG than the 757 does to the 767, and the FAA accepted a common type rating for the latter pair, the whole thing might have been an overreaction to some early flight test feedback that wasn’t totally conclusive
Given truck owners’ propensity to drill into… literally everything, the battery being the floor certainly rings alarm bells. I don’t think that would be specific to the Cybertruck though and that will make for a interesting youtube video/Just Rolled In when it happens.
My brother works in a facility that installs equipment in police and fire vehicles, they had about 10 Ford Explorers where they used the diagram to install some parts, found Ford had moved the transmission computer and they drilled into every one of them. This was in 21 so they sat for 18 months or so waiting for parts
This is the Titan submersible of EVs isn’t it?
Shop manual for battery replacement:
Step 1, remove front and rear passenger seats. (scratch head, yep, this is gunna be fun)…
Reminds me of when I first read I had to remove the front tires and wheel well liners on a Malibu to replace the headlamp bulbs.
I sold that car the next time a bulb burned out.
Those are the approaches I’ve seen more in the affordable Korean cars. Where you end up replacing entire assemblies instead of a part. It’s great for cost cutting and efficiency, I’m sure. For me (oddball that fixes my own cars as their second or third owner), it’s a nightmare.
I thought my wife’s 2004 Accord needed an oil pan gasket. Honda service manual Step 1: Remove engine.
Luck for me it was only a front main oil seal. $175 at the dealer.
Honda lost focus on that model Accord. I had one from a few years later and replacing the clutch basically required removing the engine too. Made it much more expensive than the prev model
So I’d be sitting in a pretty boring interior, on top of a huge battery, with no physical connection to the steering rack, in a car that looks like it’s still being rendered in a video game?
Look I’m not going to shame anyone for what they like but this, this I just don’t get at all.
Hey, some of us are into the low-poly pause-rendering-job aesthetic. That said, I just want a CT without the “Teslaness”. Meaning I want a CT I get into with a physical key, that has analog instrument gauges (digital dashes are fine, not screens with graphics), and door handles. And crank windows. Actually someone just get me a re-bodied 1993 Ford Ranger please.
Like I said, no shame, I just personally don’t get the appeal. If they actually built an electric car with all the ‘basic’ features and none of the weird future vibes I think they’d probably sell a bunch more.
I think lower priced Chinese models are like that, but everything above the basic tier is still all about tech for the sake of tech regardless of practical benefit to the driver
I realize batteries underlying the passenger cabin is common practice in EV design, but with the battery casing serving as the floor in the Cybertruck, I’m wondering about thermal propagation in the cabin in the unfortunate (and unlikely) event of a battery cascade failure.
I know there are all kinds of designed in barriers within and between battery cells as well as cooling systems to prevent cascade failures, but they can still happen and horrific fires have sometimes resulted.
In the event of a catastrophic failure, say from physical damage to a cell or cells, I’m concerned as to how quickly temperature might escalate in the passenger cabin and how fast toxic fumes (perhaps from melting carpet?) and fire could inundate the cabin.
Engineers have obviously considered these scenarios and designed in mitigation systems, but it does seem likely that sitting on top of the actual battery would be riskier, should a cascade failure occur, than having another floor layer between you and a runaway battery.
Hope I’m mistaken.
Those cascading failures typically take awhile to get going, so passengers really should have plenty of time to exit the vehicle, unless they’re somehow trapped
Yeah, I suppose a collision that traps passengers and initiates a cascade failure is the worst case example.
I would be concerned about being trapped. After a wreck, it’s going to be damn hard to find that ball bearing rolling around in there to break out one of the windows.
Good thing they’re not bulletproof.
*Reads about the billionaire lady who drowned in her Tesla*
Yeah, uh, somehow…
Any car is tough for a drunk person to escape when going under water. Windows are confronted by 12v battery, whether EV or ICE
It will be fine unless the door releases are all electrically operated… oh… wait…
They have mechanical releases on the door. The only time you may end up trapped is if you’re in a lake, and gas cars trap you there too
I don’t see much difference if there was a tub over the battery or the battery was reinforced like it is here. Seems like 6 of 1, 1/2 dozen of the other.
Just wait a few years until the road-salting states start having CTs with random stuff breaking underneath. That will be interesting, but possibly tragic for the owners, their passengers and the other road-going parties involved.
Either you have the battery lid + the floor metal, or you have just the battery lid that’s probably twice as thick and strong, and fireproof.
Not sure it matters.
I found the perimeter seal all the way around the battery to be concerning. If you have a bolt in floor, you have to have a way to keep water out. What is the life expectancy in the real world for that floor seal? It might be just fine, but I imagine 20 years from now, an Arizona Cybertruck will just be full of road dust and wind noise full time, like keeping the knee vents open in a 60s car.
I suspect many older pickups are like this, but 1980-1996 Ford pickups have large access panels bolted into the floor. They’re sealed with some foamy rubbery goop. Some of these are now 46 years old and in my experience they typically maintain an excellent seal, and are quite difficult to remove because they’re glued in really well.
If Tesla just does what Ford did, I would expect a lifespan of 46 years minimum.
You make a fair point if it is glued/foamed. I was working off the pic of the floorpan/battery, where there is a 1 inch-ish wide black line going all the way around the battery, not quite at its perimeter. It looks like it might be a rubber gasket material. That is however, an assumption based on one photo, but since this is the internet THERE IS ABSOLUTELY NO WAY I COULD HAVE MADE A MISTAKE!!!!! In the real world though, I could be totally wrong about that.
Kind of curious if that steering feedback motor is direct-drive like on a high-end simulator rig, or if they’re using some sort of gear, belt, or chain drive. Either way I find it kind of amusing that force-feedback motors have now made their way from the simulators into real cars.
Dodge is going above and beyond with the sim tech in real cars, with their proposed vibraty system in the EV Chargers.
The funny part is that you called this a real car
I’m still not sold on steer-by-wire. Wiring issues are my bread & butter up in the land of salt & snow, mechanical systems have a much lower failure rate than electronics.
In my curmudgeon opinion, steering and stopping should be mechanical systems in road going vehicles.
Sure, these systems are used in a great many industries. But planes and boats don’t have pedestrians and people performing lane changes without signaling.
And all safety critical “by-wire” systems should have true redundancy. That doesn’t mean having two identical motors operating on the same system when one would be enough. It means having two or more totally isolated and (ideally) uniquely functioning systems each capable of doing the job.
From what I understand this is what the CT system does, as each steering motor is something like 3-5hp and more than capable of powering the steering system alone, and each has separate communication and power, meaning they are about as isolated as possible. The only possible way to make them more isolated is two independent steering racks, which nobody is calling for.
Aircraft have two separate systems for the controls, this should too but I suspect Tesla may have cut some corners
Strongly agree. Steering and braking should have a mechanical redundancy built in. I don’t care how good you think your electrical engineering is, there has to be a failsafe mechanical connection. Period.
Does that only apply to cars? Because there are literal planes and fighter jets in the sky with only by wire controls and no one bats an eye.
You know, prior to all the recent Boeing issues I would have said that it’s because planes are maintained to a higher standard than your typical motorist. Now, I’m not so sure.
I guess you don’t like brakes, which like never ever have a failsafe mechanical connection?
Personally, I feel the handbrake should always be physically connected to the braking system.
That’s true, the handbrake is mechanical a lot of the time, but it works on only the rear wheels(next to useless on vehicles like my pickup) and doesn’t engage all that hard. It might be a slightly better way of slowing down the car than just downshifting, but not by much.
Which is why it bothers me when people call it the emergency brake. Because that’s not what it’s meant for, and it’s not especially good at it.
You mean dual circuit hydraulic brakes? Which being analog almost never fail in a completely on/off manner.
They are also fully mechanical on each end, with hydraulic fluid simply replacing the mechanical link. This is vastly different than a massively complex system of wires, magnets, and sensors, all of which need to be in good shape and communicating smoothly with each other through the cars CAN bus.
Right. Hydraulic fluid replacing the mechanical link, but there is no mechanical link. I’m not saying that brake by wire is a good idea, but mechanical redundancy is absolutely not the norm in braking systems.
Again, it is analog system, and a very simple one with over a century of use and refinement – and even then it has dual circuits. To pretend this is in any way the same as a largely untested massively complex electrical system with far more possible failure points and an on/off failure mode is to be disingenuous.
You seem to have had a hard time understanding my comment where I made it clear that I was not advocating for brake by wire, only noting that conventional braking systems do not use a mechanical link. The only way that hydraulic and electronic braking are similar, and the only way I indicated that they were similar, is that they both lack a mechanical link.
And if foreign hacking expertise gets any better, they could issue a malicious over-the-air “update” to suddenly disable all Tesla braking and steering, resulting in hundreds of simultaneous accidents, a few deaths, and snarled traffic all within a minute or so.
Planes carry like 300 people and land doing 155mph. Forgive me for not seeing how this is lower-stakes than people who don’t use their blinkers.
I wouldn’t be surprised if down the road we learn that it’s a safer system.
You gonna advocate that drivers have the same requirements as pilots to keep their legal privilege to drive? That’s a large part of this. “training for when things go wrong”.
Above and beyond that, an airborne system failure is going to give you more time to react than your steering conking out at 70+mph as you enter a curve.
The very nature of these systems is that there is no need for trouble shooting – the redundancy doesn’t permit for temporary failure. If steering system A fails, steering system B takes over.
But it does seem like your argument equates used shit boxes in a crowded road space to being equal to airliners with extensive required maintenance and oversight. What level of salt abuse are the airliners exposed to? And 20 years later, how many of those airliners are still in use with no one having even looked at these systems in the last 3 years? Just because a tech is sound and effective in one industry that does not immediately equate to viability in another industry, though the tech does the same job.
There are all kinds of systems on modern vehicles that prevent it from driving if a fault is identified. Ever try to turn a vehicle on after an airbag goes off? For most makes, it won’t. Why? Because computers work.
Technology is going to move forward with or without you.
No one is talking about if the system fails while the car is parked. What good is a computer shutting down a vehicle for a fault when the vehicle is in motion? If its the steering system that fails the computer can’t even pull you over to the side of the road or avoid an obstacle for you.
Technology may advance, but that doesn’t mean it should do so unregulated or without commentary by those exposed to its consequences. Please see all commentary on Tesla Full Self Driving for another example of reasonable concern with unregulated tech in the auto industry.
Tell that to the guy had had a total steering failure in his cyberpunk loaded up with his family.
I guess some people believe everything they read on twitter. That claim has not been independently verified by anybody.
I don’t use twitter, but I do know if it is true, it would be very on brand for Tesla.
Airplanes aren’t subject to the slow, steady and pernicious damages caused by road salt.
Yes, road salt exists. Your regular steering doesn’t turn to mush so I don’t know why this one would.
And planes that use FBW need them. Not only would it be impossible to mechanically operate the flight controls due to the forces involved, aircraft like fighters require finer, more constant inputs than a pilot can deliver. There is not a single advantage to DRW in a road vehicle beyond maybe easier implementation of some wuss BS like adjustable steering effort (still possible with POS EPAS, which can also get stuffed) and saving the OEM some money when trying to package the complicated drivetrains . . . wait a minute, this an EV with much easier packaging! Maybe there’s a safety argument, but the driver’s seat has traditionally been the safest one in the car, mechanical links and all (at least since the advent of collapsible columns), so . . . nope. At this point, if Tesla changed something I had previously thought needed to be changed, I’d reconsider my position.
It can’t be cheaper either. We’ve tripled the number of electric motors in place of the steering shaft! I can’t imagine how ungodly expensive that rack will be to replace. Compared to the manual rack in my Sorento, with the EPA mounted under the dash and out of the weather.
I don’t understand how it’s cheaper, either, but I also can’t figure another reason to do it—potential liability, no real steering feel, potentially turning off educated customers (every single person I’ve talked about SBW with is absolutely appalled by the idea and there has been a lot of them I’ve talked to since Nissan started it and at least Nissan had mechanical backup) all for something EPAS and a column can already do. If I were head of an OEM, I’d use that my competitor had SBW in ads while touting our mechanical link, perhaps turning up the scare volume on SBW, but I think it’s quite a valid concern and would rightfully give a lot of people pause if they only knew it existed. OK, not Cybertruck nitwits, but normalish people who have lived their lives surrounded by failure prone/shorter lifespan electronics of often disappointing function. If this system loses power, it’s going to be sudden and complete. If mechanical steering fails, there’s generally going to be a lot more warning and likely still some bit of control in most circumstances. Plus, this SBW is still subject to the same mechanical failures with the sole exception of the column.
This is really only true for power train parts that generate enough heat to evaporate any moisture on them at a fairly high rate. As the use of cast aluminum for body structures increases, you will see more and more of this protection. The problem is galvanic corrosion, and aluminum happens to be one of the worst affected metals for galvanic attack. Since most of the Cybertruck has stainless steel paneling and frame structures, any moisture that gets trapped has a high likelihood of forming a galvanic cell between aluminum and steel parts that will quickly corrode the aluminum. Cast parts are especially vulnerable as they tend to have larger grain structures than plate or forgings which increases the available Gibbs free energy for corrosive attack.
The marine and aerospace industries have developed standards and practices for stopping this sort of corrosion, and the usual solution is multiple layers of corrosion inhibiting compounds on any potential (heh) members of a galvanic cell.
I was surprised these folks didn’t know. It is fairly standard in multi material coatings and corrosion resistance methods.
For some reason aluminum (as well as titanium) seems to have a large body of “common knowledge” surrounding it’s applications that simply isn’t true- like the statement that aluminum forms a tough oxide layer that prevents further corrosion. In normal atmospheric conditions that naturally occurring oxide layer is around 4 nanometers thick, which is like 13 whole aluminum atoms wide. Not quite gold leaf, but pretty close. Annodization can up that by a factor of 1000-2000, but even then your 4-8 micrometer thick oxide layer is 1/10 the thickness of a sheet of paper, and a brittle ceramic to boot.
I think previous generations did so much work with steel in building the modern world that a fairly accurate and detailed body of knowledge of it’s properties passed into the common consciousness over the centuries even amongst those without a materials background, so there is much less “bro science” about it.
Luigi Galvani strikes again! That coating is puzzling because I think I see a lighter color or the rear seat pillars, a darker color on the battery pack (or maybe just more layers, same stuff) and then minimal or no coating from what I can see on the gigacastings. Maybe GCs are just coated to keep a “raw” look? To me, the gigacastings would seem to be a lot more exposed than a rear seat pillar hiding from the wet and salt above a huge structural battery pack/floor, so why bother coating those?
The GCs are certainly coated as well, there are lots of coatings for aluminum and some of them are clear (Alodine 1500 for example) to assist with subsequent inspections and maintain visual appearances.
I would expect that all of the aluminum structural elements are coated to some degree, as you don’t need actual moisture immersion to form a cell, just condensation inside a passenger cabin is sufficient to cause havoc.
This is Musk we’re talking about. It’s entirely possible they never solved all the gigacasting’s cracking problems, and the coating is there to simply hide the fractures.
And then you’ve got stuff like Ford’s factory corrosion warranty, which specifically excludes galvanic corrosion because it won’t, or is unlikely to, perforate the sheet metal, even if it blisters off all the paint
Also looks the same colour as a finish I saw applied to magnesium as a flame barrier.
“Cast parts are especially vulnerable as they tend to have larger grain structures than plate or forgings”
This is why I consider these gigacastings to be highly susceptible to micro-fracture, leading to a critical mount failure just from normal expected loads, let alone any accident.
Modern casting technology is pretty amazing with what it can do to control grain structure, but yeah it ultimately comes down to how rigorous the QC and post-fabrication inspection regime is. If the casting process is properly designed and inspections are detailed enough to catch the crack starters, then you wouldn’t expect a failure rate any higher than for comparable forged or machined parts. That “If” is doing a lot of work though. Time will tell.
I’m not an aluminium alloy casting expert, and not privy to their miracle alloy composition, but it seems quite the stretch to expect those molded in crumple zones in the rear to behave as intended. Frankly it seems like this whole program is pushing out a lot of unproven innovation. All for innovation, but it needs thorough testing/evaluation Before Release. Certainly hope that was adequately accomplished, but haven’t seen evidence of that, and the company track record in that regard is suspect.
So…. Aluminum isn’t just corrosion resistant on hot power train parts. Aluminum control arms and wheels have been getting blasted with salt and sand on most cars for several decades now.
I do imagine that corrosion could become a real issue in tight spaces between the stainless and aluminum.
48v and rear wheel steering are not new to the cybertruck but Tesla did extend their use. Steer by wire, high speed vehicle networking, stainless exoskeleton, and extensive gigacasting are new cybertruck innovations that need not be unique to EVs.
Overall, it appears that Tesla used the CT as a skunkworks innovation lab to try out new technologies that allow new capabilities while lowering build costs. Some might dislike Musk or his vehicles but he is definitely driving change in the ossified auto industry. It remains to be seen if it is for good or bad.
Steer-by-wire has been around since 2013’s Nissan Infinity Q50. It’s not new for the CT.
Didn’t that system have a real steering shaft as a backup, though?
As a much needed backup, yes.
Tesla is first to do it with steering wheel completely disconnected from rack using only redundant sensors and motors.
Which seems very unsafe to me. Steering is pretty important while driving.
I never said it was a good idea. Then again, I felt the same way about brake by wire.
As far as I am aware, the system is triple redundant, meaning there are at least 3 sensors on separate communication busses that all account for the steering system, such that should one sensor become faulty, the other two can correlate and override the faulty sensor and guarantee system accuracy and safety.
While Tesla (and Elon) are not my favorites, they seem to be taking the safety and reliability of their core operating system extremely seriously. It’s also important to mention that this has been a system in use in the aviation industry for decades, and with a similar (if not equal) redundancy setup, and as far as I am aware there has never been a failure ever.
The system isn’t truly redundant in the way fly-by-wire systems are. I believe this is something that was covered on the Philip Koopman episode of The Smoking Tire. Basically, a truly redundant system has multiple isolated and often unique mechanisms for completing the same task. While the setup in the CT may insulate against an errant failure of one motor or sensor, it does not account for larger design or manufacturing errors in the same way aviation redundancies would.
Knowing the “wing and a prayer” attitude Tesla has had towards Autopilot and FSD, I wouldn’t trust Tesla to do their due diligence to ensure steer-by-wire is any more reliable.
I agree. I simply do not trust Tesla to do the right thing. Ever.
I was hesitant to bring up aviation for this reason. Even if a whole A320 goes black, there are still cables to operate the horizontal and rudder for pitch, roll and yaw control. What happens if a Cybertruck goes dark and loses all power? You just get yeeted into a wall?
Every single one of these things requires electricity, which is a single point of failure. Not redundant.
This is no different from an EPAS rack on a modern car, if it is push-button start as most cars are now, should power go out, power steering is lost, and the wheel will lock. Redundancy doesn’t solve every possible failure mode, but typically is makes it such that should that system fail, something worse has happened.
The same goes for aircraft Fly By Wire, it needs power too, should power be out, you’ve got bigger fish to fry, and the FBW system would not be the cause of the resulting accident, and the same principle applies here
well that sucks
If the power fails you loose power steering but it shouldn’t lock the rack, and the steering wheel lock itself should not engage unless the car is in park.
Aviation systems are also maintained by highly trained mechanics and have aggressive maintenance and part use lifetime standards. This was about making the front structure design easier and tech talking points, safety considerations were at best third.