Hi everyone. Welcome to Monro Live. We've got an exciting series or episode that we're going to stitch together here. I'm with Lars, Eric, and Hector, all from the Tesla team. I'm sure you know these guys behind us. Obviously, we've got the Cyber Cab. So, we're going to dive into the unboxing. We're going to walk a bit of the line. We're going to look at batteries, but it's going to be exciting. I don't know if it's going to be more exciting than the Roadster video from...
Wrong hat. I wore the wrong hat. Everyone's got a cyber hat.
We got to swap hats. But so I mean, let's dive into what we're seeing on the floor here. Like, what was the design intent? Like, the whole unboxing discussion came out a while ago?
It did. I think we first kind of—it was probably like five or six years ago. Yeah, we were working on like automating our lines more and more and more, and we kind of hit a wall where like the problem wasn't the automation, the problem was the design. And you know, my team is responsible for not just the design of the vehicle, not just the test of the vehicle, but also the manufacturing and even the automation groups that we have within our team. They're all in one engineering org. So, we got a few of us together and we're like, how are we going to break this? And like that was the answer. We had to break it apart. And you know, Eric at that time was in charge of our integration team, and that was a big task of like a hundred years of building cars one way—how do you break it apart? And so we kind of looked at a car and realized it was a box, you know, there's like a bottom, a top, a left, a right, a front, and a back. And then we came up with the idea to unbox it, and really that did two things for us fundamentally. One, it gave us smaller pieces to work with, so we could get robots in tighter, and we'll walk through each bit a little bit. And then two, it also allowed us to break the line apart and go to more like a tributary and main flow. And so, you know, the manufacturing guys like Hector works with, they're able to make final line. How many stations is like 25 stations?
25 stations fully automated.
Like final GA when the whole thing comes together is 25.
Wow. And like a normal Model Y was probably like 5, 10x that. Yeah.
Depending on which factory and how much you're running. So...
Those are the two big things. But why don't we start like piece by piece. Eric and I can give you a quick review, and then Hector will just tell you how we put it all together.
Why don't we start up front, Eric?
So the front—yeah, the front is one of the more obviously dense and integrated modules that we have.
This is the business unit.
This—yeah, exactly. This is basically where, you know, 50% of the parts and the content really is at. And we did that on purpose. We wanted to have everything isolated to one area for efficiency of low voltage distribution. We put all of the endpoints, all the electronics where the control systems are. And so that's kind of what we've done here. And we've also revolutionized, which we'll talk about a little bit later, how we do thermal system management. Moving, you know, another step from our super manifold of previous products. We've got a 48-volt low-voltage battery. And then on top of that, we've got the drive unit, which is basically right in here in the center. And one thing that we heard a lot from our service teams really early on in this project is we wanted to make the drive inverter serviceable without having to basically tear the car from the underside out.
And so what we've been able to do is allow this assembly to be very simple to remove. Obviously the washer valve is very simple to remove. It allows you access to the inverter. So we don't even need to raise the car, tear the suspension, tear the drive unit out to service the inverter. And so the fundamental aspect of integration here is principally insane.
And you can't like do that without thinking at the beginning, because Eric just very casually said, 'Oh, we just take out the thermal system.' But it's a refrigerant system, right? You can't break refrigerant lines, right?
So we actually flipped it. It's a coolant-based system for heat exchange, and the refrigerant system is completely closed.
It's a closed-loop system within the...
Which you couldn't worry about after the fact, to your point, right? You had to do it at the beginning.
And so like it went all the way back to the beginning of design, and then you know, to drain and purge a coolant system is actually pretty easy.
And we even on that, we actually made a through loop so you can push and pull and do it faster.
Oh, that's brilliant. Suck it out. The things that really stand out to me—I mean, we've obviously torn down a bunch of the other Teslas. It's the packaging and how tight the drive unit is in Z. Like it's way higher in Z relative to the super beam. And actually, you know, so what is mounted here? Is this the EAC compressor?
This is the AC compressor and basically the coolant manifold and the coolant controller all one assembly.
It's all integrated together.
And that goes back to the principle of integration and test. We talked about that tributary flow. Everything is self-tested before it goes in. So when Hector is building it, he knows he has a good part, right? And so we have a module line that builds the integrated thermal or the indirect thermal system, and that's then self-tested on the line. So we know we have a good part coming in, and you'll kind of see this like every part comes together. Same with the drive unit. And then by the time we get here, you have the full front end that Eric was describing with all of this. We actually know the car is ready to drive and go, and all of the computers are hooked up when the front end comes to line. So it's pretty cool. But we really took a lot of the principle of moving complexity out and building modules that bring assembly upstream so we can self-test more in larger modules as well as in sub-subassemblies in a much higher percentage than anything we've ever done before. The other thing I guess I'll say on the integration side, which I know you're keen on, is like this has been the core...
I am keen on it as well. This is a core of Tesla, right? Is an organizational structure of engineering that works really well together. There is no veil. There's no wall between teams. And that's kind of what you see here.
Well, I think Lars and I were actually, you know, talking on the ride over here. We were talking about the patents, the designs, you know, what is the secret sauce to Tesla? And I think you said something like, well, it's actually not that secret. A lot of it's pretty public. And I'm like, I think the secret's the culture. I think it's what you just touched on. I think it's this mindset that, yeah, it's not super hierarchical.
It's our problem. It's not your problem because you're the thermal guy or you're the chassis guy or whatever. And this is the net result. And by the way, I think this is the first example that I've seen from Tesla where the EAC compressor is PIA or part of the assembly.
No separate bracket. You know, sometimes unintended consequences.
Yeah, like we actually, you know, we put it all together, we're like, 'Oh man, this is way heavier than we anticipated.' And we had to like refigure out how to isolate it because it actually added to the modal mapping in the front where normally when it's sort of just a tiny little compressor, it's not that big of a deal.
Well, there's some like really cool stuff here, right? Like we are no longer running high-voltage distribution from the battery pack to the system. We basically do a junction off of the drive inverter to power.
Ah, there's another Conway law break, right? Like...
So now the inverter team is working to make the cost of the high-voltage system and the thermal system less, means less parts, less complexity. You can ask questions, and then I'll go to the next one.
So the inverter kind of became like this little substation, like a power distribution, a PDC to go and power everything else. So you're not running dedicated circuits all the way from the battery pack.
And that is 100% the core of this car is like do not waste a gram. Do not waste anything.
Yeah, yeah, yeah, yeah. Oh, it's amazing. Okay.
Come around here. Like let's look at the suspension real quick because we got a lot to talk about.
So it's a strut suspension, which you wouldn't think with the low hood. So, we had to offset the kingpin to be in front of the center of the wheel there. And we'll talk about this a little bit more later too, but it's also our first brake-by-wire system. And again, that was done because it was almost a necessity for breaking it apart. You can't have a closed hydraulic system for braking if you're going to split the car into pieces. And so, we can self-test the front brakes and the rear brakes separately.
Well, this is beautiful, right? Because we no longer need hydraulics running from the back to the front from a booster or from some kind of junction. It's basically everything is isolated to the module that it's in. And it just again allows for us to take cost out of the distribution of whether it's conductor material or fluids or whatever it is, and allows us to simplify manufacturing substantially.
So it's allowing you, if I heard what you said correctly, you're doing testing like on the thermal management system by itself, and then you're able to plug all these components, brakes, drive unit, all that into the module, this unboxed chunk, and then you can test it in this state to ensure before it gets bolted to the rest of the modules that this is ready to go.
Got it. You're buying down risk at multiple ways every line. What's our full percentage? 85, 90?
The full line, all of it?
It's about 78% automation. We're working on the path to...
Said 80, I'm just kidding.
Round up just a little bit, that's okay.
But to do that, like you have to have super high OE, which means all the parts going in have to be perfect. And so we tested here, but even like—I'll let Hector point this out—even like the thought process of how we were going to put it on the pallet.
You want to point out some of the features underneath and how we put the subframe in?
Exactly. There were also creative ways to assemble this. As you know, the car is pretty low to the floor. So, the springs are actually angled. So, we were trying to find a way to install it in like a sandwich mode. The geo pallet actually sets the subframe in position and all the components in previously. And on the other side we get the casting with the dash and the instrument panel preassembled, and we sandwich it in. The creative thing here is that we were automating how to basically align the springs to the casting with full automation.
Well, and what's fairly unique here at a very high level, most every unibody application out there, they're decking the drive unit and/or cradle subframe to the body.
This is inverted, meaning the body comes over this.
And so you're saying that was part of the enablement. It's actually pretty amazing because you basically most of this system of...
Thermal, drive unit, suspension.
It sits on a pallet and we literally drop the body on top of it. It's one of the coolest stations we have.
Again, you kind of have to bend your mind when you're doing it. And it's very early on when you're in the design that the guys that are building the equipment, Hector's team, they have to come in and say, 'No, no, no, no, no. We're not going to go this way. We're going to come top down.' And that means like you got to change bolt directions. You got to change sequences. And like that's way at the beginning of design. They don't get an over-the-wall design to manufacture.
And in fact, in this program, I say you guys were telling the design engineers more of what they needed to change.
Yeah, there's a lot of conversations, daily conversations, just to go back and forth how it's the best way to actually manufacture the vehicle. One of the coolest things is I remember three years ago we basically rented out a space and every engineer associated with this project went and stayed there for three weeks and whiteboarded all this out. It's manufacturing, it's software, it's every vehicle system, everything. Just like literally in a room, you can't leave. You're going to figure this out. You're going to basically break the constraints that have ever existed. And that's kind of what came out of it.
It's only paradigm shifts. Exactly.
Put everyone together. Let's keep going on the rest of it because we got a lot to get through. So every module in my mind has like some kind of super function. These are the side modules, left and right. They're almost a mirror but not quite because of some distribution. But this is like for me, this is the safety module, right? Obviously frontal crash is important, but in an autonomous world hopefully that never happens because we'll stop before you run anything. But side crash is something that's very, very difficult to get out of the way.
And so this is actually a sandwiched set of boron steel door rings. It's a single piece inner and a single piece outer. And its main function is just to stop things from hitting it. Both the battery down here and then, you know, obviously the occupant space here integrated with the door. This is also put together, you know, automatedly. But it's a great thing we can do is like, you know, when you normally have to put all this stuff together, it's kind of always up and left. But like because it was its own piece, you can build it flat.
Yeah. A lot of other assembly lines have these chairs that people sit in because it's ergonomically insane to get into a car and assemble all this.
Well, and I think the same is true with your...
That's exactly what we were looking at, right? When we looked at the header coming in and you look at the overhead console all being part of that module, I'm like, how much better is it to work on it when it's on a bench at this height versus trying to reach into the center line?
It does two things. It improves like operator efficiency when we're using them, but also reduces mistakes.
Sure. That's absolutely...
And it's great for humans and it's great for automation.
Turns out whatever is good for a human actually makes it easier to automate. So exactly right. We—there's this thing at Monroe. It's called Bob the one-armed builder. He's like blindfolded, one arm, whatever. And we're like, irrespective of what your process is, if it's good for Bob, it's good for everything. And that cascades down.
So I'm curious, as we're going through the rest of this unboxing, one of the nagging questions I've got, there's statements out there. It's like, hey, 60% part count reduction, and clearly there's a throughput benefit, there's an OEE benefit, there's all these things, but is it actually a net reduction of overall floor space? Have you found that?
For sure. It's about 50%...
Compared like per unit out.
Yeah. The way to think of this is like a car that is full size. Just think of the space that requires per station for a full-size car.
What's the pitch length on the front module line?
On the front module is about two meters. So we basically use every space around the front module to work on it. So that's what this enables: automation. Fully automation can reach from different sides of the assembly. So that's what really unlocked...
And this floor space reduction is something that like from day one, the whiteboard session, you're like, this is going to enable this.
Well, it was actually one of the main targets. We need like—we're not going to... like we want to build thousands and thousands and thousands of these a week.
We can't build a giant building at the scale that you know like we need to get the output. And so like we're like, this is a space we have, it's where you need to get at least a 50% improvement in.
So seats on pack, you can see it here, but like we'll talk about the pack later. We actually did this already. We did it with YSB. We do it with Cybertruck. And that's why I think when people are like, 'Oh, it's so revolutionary,' I'm like, 'Actually, you guys haven't been paying attention. We've been doing it all along incrementally.' Yes. It's hard to change something overnight. But that, you know, the seats go on there, the carpet goes on there, it all comes up together. And you know, we kind of—this is our third iteration, so it's probably the most complete.
At this point, it's a relatively simple assembly. Like we've gotten it to that point at this stage, but it's still like a lot of part deletion. It's very simple, but there's not that many parts on it because we've generally deleted everything we can.
Most of the work—this is another thing. It's like most of the work was on the battery pack line, and then we had the seats line, and then I remember just deciding, okay, who's going to do the center module? Because it's like which line will we extend? Will the battery come to the seat line or will the seat come to the battery line? Yeah. Anyway, and then the rear end, it's pretty simple. Actually, one of the things I want to point out that like I don't think a lot of people noticed is we were actually not coating the front and rear modules, only the sides coated.
Yeah. So I was going to ask exactly that when we were up front because if you look at the Model Y or the Cybertruck, they're all e-coated, right? So what is it? Because it's a robo taxi or is it just like, hey, we actually don't need to do it.
We don't need to do it. And like it was also a point at the beginning, one of the most expensive parts of a factory is a paint line.
You know, few hundred million, million man hours to build one.
So like rule number one of Tesla: question the requirements. Why do you need paint?
And when we went, why do you need e-coat? And it turns out aluminum and our materials team did a great job with this, but like you know, the alloys we use in the castings, like they don't oxidize and reduce strength over time as quickly as other aluminum. You know, most 356 materials like on the chassis parts actually after you age them to T6, they continue to gain strength. Yeah. And then we had to pick, you know, carefully our sheet aluminum and the joining methods, but you know that part of that reduction in floor space was like we got to delete the paint shop.
Well, in fact, it may actually help with some of the bonding if it's not coated, right? You don't have to worry about that. Now, is this something that you view as backwards compatible with a Model Y or a Cybertruck?
Yeah, I mean, again, I think it had to come up front. But like certainly like as we move forward, we're going to try and put that in because it's part of the mission is like a sustainable future. And so like the less, you know, we can put into the product that is bad, the better we'll be on the back end. And this is all with doubling the useful life at a minimum of the car, like the car is going to run half a million miles at least.
Yeah. I mean, you think about a robo taxi, it's going to go more utilization every day. So...
All right. Are you running, Hector? Can we go up and take a look?
We're running framing line. So I can explain you a little bit of what we do here.
So while we're walking here, I just want to get this on the record. We're actually—we don't run every day because the line can run so fast. It's currently outpacing our deployments. So it's a bit of a race right now. Who's going to go faster? Right now the, you know, the manufacturing line's ahead. So that's why you don't see everyone all busy here. But...
Well, there's a lot of lines that wish they could say that, right?
Yeah. There are a lot of people that wish that was the problem.
So as we just saw, the product is already pretty cool, and obviously the line has to be as cool or even better. So we were looking on creative ways to build the car. Obviously, as Lars mentioned, we're doing a box architecture. So that definitely enabled a lot of the automation that we have. This line, which is the framing line, is fully automated, 100% automated. Even part feeding to the line is fully automated with AMRs. And basically what we...
AMR is an automated mobile robot. It's a trolley that goes underneath all of our material storage pallets and can move them around on its own.
The viewers are really going to appreciate that you just explained that. No, for sure.
Can I like even more so for, you know, not us.
Yeah. Yeah. Basically what we do here is what other or like what regular vehicle manufacturing does is framing the car, putting the car together. This is where actually the car looks more like a car. But in this case we have all the components, trims, harnesses, hoses, everything on, which obviously makes the things a lot easier later for our final assembly. So we're basically building from inside out and enabling obviously as we said multiple times.
So when we look at all these modules and all these subassemblies, this unbox process from a GD&T locating schema, gap and flush, is it a net improvement? What were the challenges?
I think at the beginning we all were like super afraid...
GD&T is geometric dimensioning and tolerancing.
There you go. Good job. Because like it hadn't been done before, but when you back it up to like what's the first principles, actually it is a benefit because we're building to net nominal, right? Like so by the time we get here, whatever variation's gone into the side module and the front and rear modules, it's like when you frame it, you put it to like the perfect position, and then your gap and flush is actually better because like you don't get the sag that might occur in an oven when you e-coat it, and then you don't have to worry about adding parts on and then like having each of those fall out of tolerance. It's like you build the module right, when you put it together, you frame it right, and then the last 25 stations are dealing with a much, much more accurate assembly than you would have if you kept adding it together. One of the primary examples I can give of this is like Boeing, you know, they do a stack tolerance where they build the fuselage... or I shouldn't call out Boeing, most aerospace, I don't want to like badmouth Boeing, but each section has a tolerance. And so the actual like variance of a plane is upwards of feet by the time they get to the end when you're just talking about like maybe plus or minus a quarter inch. But if you have, you know, 20 or 30 of those, it adds up over time.
Yeah, totally. That makes sense. You're essentially localizing all the GD&T on a module where it's in a better ergonomic position. You can fixture it better. Everything is superior. But then the interface points on the final line are a fraction of what they would be typical.
Exactly. The meaning. Yeah.
And we—I mean we validated this. We didn't just throw it on the line. Like we actually—this is crazy. No one knows this, but we actually tried this...
We actually started working on this on Cybertruck.
Long many years ago. We started actually validating this on Cybertruck. Now it didn't make it to production because it was kind of after the fact. We used Cybertruck as a platform to really learn many, many years ago how to get this right because the architecture of Cybertruck is actually very similar on a body structures perspective. So we actually like—and this is the power of Tesla, our automation groups started working with you guys before the design was done. Like we built framers, we built these machines to like put it together. We used Cybertruck, we used Model Ys, we used early designs of the Cyber Cab to like figure out what the tolerancing needed to be to make it all go together. Yeah. Yeah.
If you come over here, it's actually kind of elegant because this part, you can see some of the skining, but like the seal flange here is like it's not perfectly flat, but it's planar. So when we put it up to the casting, we have a nice solid seal. And the same is true on the bottom to the battery pack, but it goes this way.
And then in the back to the rear module goes that way. And that's like again, you have to do that at the beginning. Otherwise, you're never going to get that single-axis sealing or as much as possible anyway, right? And then from a locating perspective, you're just bringing this in with the header. You've really just got two main interface points, and that's going to set up very...
And then you know, you always worry about Boeing, and like it's just like put it here, bolt it down.
And that's actually what we learned from the Cybertruck is where we had to clamp the assembly to get—you really think of like framer today in a body is doing the same thing that this is doing, right? You're just moving framer into this area of GA with general assembly with all these parts on it. So what we had to learn was where to clamp, where to grip, where to hold the assembly, where to datum the assembly in that process to make sure it all is automated.
Even this whole like clamp design and how it's pushing from the inside out was thought out from the beginning.
Yeah. I mean, one thing that was very apparent when we got the Model Y and of course then the Cybertruck when we're looking at the castings is most body and whites have slip cross-car, right? Because it's a section over a section and it allows you to adjust, but it's locking everything net in the Y axis. We believe in precision first.
Why don't we go talk to some of the guys about the components that put it together?
All right, everyone. Now we're going to get to the battery pack of the Cyber Cab. So it's going to be super interesting. I'm with Robbie. He's the director of all things battery here at Tesla. So Robbie, thanks for being here. What is new and/or carryover about the Cyber Cab? Like let's talk about from its use case, right? It's not a consumer vehicle. So what did that influence in terms of the design? What are we looking at?
Yeah. So Lars probably already told you about our focus on scale. First thing I'll say is the 100% automation and 10-second cycle time target drove a really intense focus on simplification, which I'm going to talk you through some of these things. We deleted a lot of parts and we integrated a lot of the remaining things. You get two for one.
And that also helps us with reliability, huge goal, and ultimately like total cost of ownership. So battery cell and non-cell is often a quarter to a third of the cost of the entire vehicle.
Then you add reliability like service cost, replace the pack, super expensive, plus the labor to do it...
Plus the downtime in this application.
Plus the downtime, and then you know, cost of mass, this thing is super heavy, that impacts TCO. You have fast charging, the faster you charge it, the more uptime you get, the fewer posts you need to put out in the field for infrastructure.
And ultimately that all just leads to lowest cost of ownership that we've had an intense focus on with this.
So probably one of the first things you notice with the architecture is we took the ancillaries and put them in line. Right? So you have a flat vehicle pack. You can see over here.
We kept the pack top cover as the body floor.
So we can do seats on pack here. Seat crossmembers directly welded to the top cover.
Yeah.
And then you put this in line so that you don't have this bulge up here on the pack which kind of gets in the way of the seats as they recline. As we're thinking about future vehicles in a common architecture, you can have sort of the length of the pack vary and have it not impact your feet positions if there's a second or third row.
This is something we've kind of like started to push back in like the Model Y has a flatter pack for the... so you know there's a little more space for the third row feet there.
But one of the things that like you may not see at first but notice is like typically we've sort of looked at the pack as like a top-down assembly. And in this scenario when we put the top cover on, this is part—the ancillaries are sealed at the top and actually serviced from the bottom. And that was because when we looked at all our service data, the pack itself, the modules and the cells, they're actually much more reliable than the electronics behind it.
Yeah. So if there's anything that we need to service, just like we talked about with the thermal management, you purposefully change that orientation.
You just think of it as a platter. Literally there's a mounting structure on the bottom. You pull it down, kind of pulls all the ancillaries down, and you have a certain...
I'll let you get up close with the ancillaries in a second. One of the nice things about this too is...
Top cover, the ancillaries, it's part of the same stamping here.
So in 3/Y there's a wall and a top cover. Those are both gone...
And a seal and a bunch of, you know, bolts to put it together. Yeah.
I'm also not seeing, unless I'm missing it, you know on the Model Y there was a bunch of tailor welded blanks like laser seams on that top panel. I don't see those here.
Yeah, those were the—so it was up here. That was the wall for the top cover to create this volume there. So that's gone. Now...
You'll notice too it's steel top cover. We always do it. It's a steel enclosure base.
That's something that is a first on this...
program. And there's a couple advantages to it. One is just cost, right? Steel $1.90 per kilogram versus 5 to $8 a kilogram. There's a cost of mass too where you lose on steel, but it still nets out. And then one of the things that you'll notice, it may not be super obvious though, there's actually more parts on the enclosure. It's one of the to drive this thickness down to 1.2 mm.
You need to localize some thickness.
You need to add a bunch of local features and strengthening around it.
We're still learning whether we think this in the end was the right call, but it has its advantages of being lower cost and mass.
How much from a cross-section perspective on the pack? Like the Model Y, at least in three dimensions, was structural to the surface of the pack, but on the bottom it was open, the four after it was that structural foam didn't actually go to the outer bounds. So from a structures perspective, how much structural work is the pack doing on this versus like a Model Y?
Yeah. So in that particular, see that better over here, but yeah, you have durability, front crash, side pole. Those are your three big ones, right? You were talking about side pole. Side pole. So what the pack does, you don't want the cells to absorb energy. You want them to be stiff backup structure to help the main absorption structure stay put. Right.
So inside pole and a structural pack, that top cover that the crossmember is mounted to, that giant surface area of adhesive really helps stabilize it. Here, that top cover actually fastens into the main enclosure through these longitudinal. You can see the through holes here. And so you're not sending load directly into this adhesive cell plane. And I can energy skipping over to the beam and then going to the next one over.
But something cool we were able to do with this is when that longitudinal comes through here, the second module on the struck side can actually resist the motion of that longitudinal, contributing to the backup structure and stiffening that. And so we basically have shear planes between each of the center channels using the module. And it doesn't crush the cells because it's so distributed. And so the module is like a stiff, you know, element there that's adding strength. And we call it, so it's kind of semi-structural in that regard.
Semi-structural. Yeah. Yeah. And when you're dispersing it over such a big area, it can actually do a lot, I would imagine.
And then I'll do frontal crash. But you know, so how do we carry the loads through there? The casting comes through here. It has a triangular piece that bolts up to the front here. And then you see these three big loops here. Those are like our main sort of load paths as they come through. So the casting will distribute it and kind of put it into five longitudinals. The two side cells, one, two, three in the middle. So you get five overall and we kind of net it out and carry it the same way through the pack to the back without damaging the cells.
Now is this from an impact perspective like the furthest outboard either for side to side that we could get them? Meaning we needed this for free crush space.
Uh no, we were way more efficient than we thought we needed as many batteries as we thought we did. [laughter]
That's right. That's right. So this, yeah, if we want to get into that, this is an 88S 6P battery pack. We have room to add a few more cells as you can see there. And the other thing you can't really see is there's space. There's more space than we need below these modules. So, we protected this pack to go to an even lower energy dense chemistry as we scale and need to bring LFP into the platform. And they still vent down and out.
The cells vent down and out. Yep.
So, I got to ask the foam and the potting question because we tore apart the Model Y and we had to use a dry ice blaster and we ripped like, you know, ratchet straps off holding tables trying to peel that steel back. The steel was ripping. This is a pretty big change where you're just potting locally in the modules.
This is water soluble. Just I'm kidding. [laughter] No, it's the same potting, but we're doing modules instead of packs.
You probably want to know why.
I do want to know why for sure. All right. So, it's good to be self-critical. It's one of our cultural attributes here.
It's okay to look at yourself and I didn't do a good job. I'm going to give our Model Y structural battery a grade D. Okay. To the customer it was an awesome product but internally just sort of like how hard it was to build and how much it costs.
Cybertruck structural pack, we made some big changes. I'd call it maybe a B minus.
I was going to give you a C+ but okay.
It was definitely passing. It was for sure passing grade. You've seen the pack. You've seen the line. Like I'm proud of that.
What we started to realize is like in the limit the grade A structural pack. Yeah. And the grade A, we'll call this non-structural semi-structural pack. Everything we could tell is like pretty equivalent cost for you're going to get maybe 10% more volume for cells in a structural pack when you delete the structure, but you're going to save a little cost as a result, but the issue is the service cost, right? And like you have to get just infinitely low cell reliability to win on structural packs because we get to salvage three of these modules. Like if this worth, you know, 1,500 bucks a piece if one goes down on a Model Y, you're replacing the entire thing.
So like as a thought experiment, a Megapack at three and a half to 5 megawatt, you're never going to pot that whole thing. If one cell goes down, you don't want to lose that asset. An Optimus battery, it's one module. It's two and a half kilowatt hours. Three, you know, it's that's kind of okay. It turns out in that product, it's actually not the largest cost at all.
And that too. Yeah, exactly.
So we found that it was going to be too much engineering drag to maintain two architectures in parallel. Let's just like go all in on making this the best it can be.
Yeah. So whether it's on the engineering design and development side or I would say the economics of service and getting to it, are we scrapping a whole pack or just a module?
So that's really what drove us here.
Exactly. So what grade is this one? I got to know the grade of this one.
I'm feeling pretty good about this one. I actually think this is Robbie and the whole battery scene like most elegantly best design pack most integrated in the vehicle.
I can't ever say you like perfect, but I'd give it an A minus.
Always a little room to improve.
Yeah, I mean that's a good grade from where I sit. So, it's also worth noting when you think of TCO as we've talked about, replacing an entire pack before half a million miles is a killer on cost. So we have to think of reman, we have to think of like pack serviceability, we have to think of all those things that actually get us to our cost target for the product.
Even maybe the insurance side of it, I would imagine, right? If you're going to insure a product like this and Tesla self-insures and does all sorts of things, that's got to be a benefit.
Anything else you want to see on this?
No, this is great. Thank you for taking us through.
Can I show you just a couple more things?
Yes, we can. Anything you want.
Okay. There's some stuff in here I'm really excited about that I think you'll like. So, one thing you'll notice is there's no harness from the BMBBS.
Yep. It's all BMBB is battery management board to the penthouse that what we call the inline ancillaries.
We weren't the first to play around with wireless, but we implemented wireless communications in here.
And it's just been huge. It allowed us to delete harnesses which are super messy to automate and money intraac wireless communications.
Yeah. So, normally this is a dumb circuit board that sends data to the HBC through some communication wires through a harness BMS battery management system. You have to package pay for it, it fails.
So now we have an RF chip on here. It drives the cost up a little of this circuit board. But now we get, you know, to delete that harness and now we have edge compute here. So we can do local things like runaway detection, redundancy within the like because each board is, you know, not now singularly powered but also like when you think about cost as I said is the whole vehicle even if your board cost went up 20 or 30 cents for the communication chip we deleted a harness that might have been a dollar you deleted the assembly of the harness might have been 50 cents and on the system level it's cheaper.
It's huge. My overarching takeaway when I look at this pack, you know, when I actually stare at it for a second, my question when we were walking up to it that I didn't ask, I'm like, are all the pieces in here or not? [laughter] Right. And that's a good grade, right? That's why we get a good grade because it looks like it's missing something.
We're missing potting over here.
So you can see our 4680s. And then let's take a look at the ancillary back here.