Inside the Tesla Semi: Dan Priestley on Engineering, Production & What’s Next

Uh, Hello everyone. Welcome back to Mineral Live. I'm Kevin. I'm here with

Mineral Live. I'm Kevin. I'm here with Eric and Dan, the program director for the Tesla Semi. Um, we're super excited to be here. I know, uh, you uh, you know, your intent is really to showcase not only the product, but the

factory and, you know, that you can, you know, build and meet demand. Just while

we have you here, can we like high level talk about like maybe some some production milestones and and program time like for Monroe and what we do a lot that's that's critical, right?

you know our design clean sating iterations going back through it and then obviously tying that constantly to manufacturing. So

manufacturing. So yeah, if can we could I imagine I think that's kind of where the story is. A lot of people have asked me this on our way here, but uh if you want to expand a little bit on that, you know,

we first unveiled it in 2017 and then as a company went through a lot of growing pains and then we had the COVID pandemic and a lot of that resulted in some fits and starts on semi- overall, but then we

were able to launch a pilot fleet at the very end of 2022. And over the few years that those trucks been running about 17.5 million miles at this point, you'll probably see some of them running around here today. it. We've learned so much.

here today. it. We've learned so much.

There's a ton of data and they've been in a variety of applications and we've really distilled all of that now into a design that's for high volume and we learned a lot and we've made some minor changes. Uh you might even see some

changes. Uh you might even see some aesthetically like we now have a drop glass.

change the front end a little bit, but there's a it's easier to look at the parts that we didn't change than it is to the parts that we did change because ultimately we rethought everything and

said, "Okay, how do we make it to be lowerc cost manufacturer so we can actually scale this thing? How do we make it that it's even more reliable?" I

mean, the current fleet's great, but we want to take it one more step and then ultimately we want to provide the lowest total cost for our customers, right?

That's about it. and reliability and volume go with that. But it's also about rethinking any design or taking areas that we already have at high volume. So

for example, we've gone to our in-house 4680 battery cell. So we went away from a purchase 2170. That's an opportunity for cost. We brought over key

for cost. We brought over key componentry and innovations that went into Cybertruck. So the stator is an

into Cybertruck. So the stator is an example of that inverter. Um as well as the whole steering system. It's not a direct carryover, but there's a lot of core componentry, low voltage communications architecture. everything

communications architecture. everything was in mind and taking advantage of all of the iteration and R&D work that had been done for years on other platforms and have a chance to say okay let's

bring this into the present day and set it up for a long-term lowcost high reliability future and that's like really what the last several years have been about taking all those field learnings plus other development

activities putting it together into one package that we're confident is here to meet the demand and now we have the factory to help make all of them.

Gotcha. Yeah. Because I think you're probably one of the few companies that has production vehicles and is also building something in the kind of class A over the road, you know, threshold.

And a lot of like the requirements that you'll see on smaller vehicles or even, you know, like one ton trucks, things of that nature don't necessarily correlate to something like this just given I think customer use cases like this one has a very different one. It's it's here

to make money. It's here to be loaded.

You generally don't want them to be unloaded, right? You know, so can you

unloaded, right? You know, so can you what did change? Let's kind of start with some of the stuff you guys decided to like revisit.

Well, I think actually in that mind a great example is the thermal system.

Okay.

Total rearchitecture. So before we had actually used two versions of the model 3 Y thermal system. We had actually taken uh two compressors and two what we call super manifolds and put them into

the prior version of the truck.

They were not at the right voltage. They

were at a lower they were at 400 volt architecture, not a thousand volts. We

had to convert that down. um you know, all of those extra components. It's

difficult to package, but then you've got all of the hosing that goes with it.

And we've now changed over to what we call the indirect thermal system. This

is our latest and greatest and it's now a assembly that it's actually there's no refrigerant lines anywhere in the truck.

The refrigerant is fully contained within the IT is what we call it, the integrated thermal system or indirect rather, indirect thermal system. And

that's built at our factory in Texas and is shared with CyberCap. So to like put exactly what you talked about is we just uh took the thermal system out of our

most efficient vehicle ever and are now putting it in the heavy duty vehicle.

These have an order of magnitude different in terms of watt hour per mile and energy to move down the road. But we

can use the same the same thermal system for both because there's that much capability in that device as well as then we pull some other things like we increase the manifold and some of the the heat exchange capability

but the core componentry is the same.

the pumps are the same. So even though that yes, the hosing is obviously different. It has to go the vehicle's

different. It has to go the vehicle's different. The radiator is a larger size

different. The radiator is a larger size because we have a higher heat rejection requirement, but those core components, they're going to be made in the millions a year. And so we have massive amounts

a year. And so we have massive amounts of reliability data from the field.

We've got a robust supply chain. There's

never going to be a shortage of parts.

So all of this goes together into into that thesis again of high reliability, low cost, and leveraging our other components. But a thermal system is a

components. But a thermal system is a good example of where we just clean slated everything and went for a much better more efficient system.

That's interesting. So, and you have two pack sizes here. Are they aside from routing, is there any major differences between essentially the two module pack and a three module?

So, the um the packs themselves are actually the same. So, in the long range truck, the 500 mile range truck, there are three packs and then in the standard range truck, they're two packs.

Okay. But in terms of energy amount and sales, they're the same.

Okay, so that's great because that means we can build them all on one line and we have one supply base and service is great because it's interchangeable in the field as well. And we tried to make this a modular platform so that we can

serve those different applications very efficiently. And so um there are on the

efficiently. And so um there are on the longrange truck for example, it's these components here where there's the storage door and this extra set of uh steps for example. But in the standard range truck, those go away and

then the whole truck shrinks. And in

that shrink, we talk about components that are designed to change length very easily. Meaning these are extruded

easily. Meaning these are extruded lines. These are hoses. These are uh you

lines. These are hoses. These are uh you know rolled frame rails. These are

things where you don't need separate tooling just to make something that is a different length.

Yeah. Just a variation, right? And so you're making something

right? And so you're making something again designed for high volume with variance so that it's easy for the factory. It's easy for the suppliers and

factory. It's easy for the suppliers and the supply chain and it allows us to make something again low cost because the truth is that these two vehicles serve different purposes in the market and we saw that we really needed to be

able to serve both of them effectively and having a vehicle platform that we can flex like that is critical.

Okay. Yeah. I mean it's you know the capacity that you're talking about here you know 50,000 units that's roughly 20% of the market share.

20% of the market in the US.

We're also I mean keep in mind we're going to North America so we'll be Canada and Mexico as well but we're also looking to start production here to support Europe as well.

So we have plenty of capacity to start we have bigger dreams than just this 50,000 and we'll get there but um yeah we you only get to the cost targets that we have and the ability to truly turn

over the fleet and get away from diesel and mass by going to volumes like this because you have to drive down the cost.

So we look at that as a necessity and it's up to us to go make sure that everything is in line both to make these vehicles effectively but then do so at a cost that the volume is not the question

and particularly with the lower cost per mile that electric provides. We expect

that fleets will prove this out understand how it fits into their operation and then will want to order a very high volume of them because it just becomes a spreadsheet and a calculation at that point. Um, and so that's really

what the economics show and we know and have been proving out over the last few years and now it's about making sure all of those customers out there uh go to scale as well.

Nice. So this plant, how much capacity you think you could do here like in the future?

So this plant is designed for 50,000 a year. Yeah. So, a thousand a week. And,

year. Yeah. So, a thousand a week. And,

you know, we are going to be doing some small expansions because as we grow, we've got some projects to uh get the most out of the factory. But, you know, this is our our first one, but we'll go wherever we need to to stand up in

another capacity when we ever need it.

Now, um I was joking with Eric, as soon as I put on this shirt, I got asked like five questions at the hotel. They're

like, "Ask about the camper." And I'm like, "I think I I know why it's not launching with it and maybe why that's not necessarily a focus point." Um, but

is it in your mind just where where Tesla sees the industry going? Why maybe

like it looks like it's package protected for it just from everything I've seen if you wanted to, right? And

maybe well into the future if you start getting into those more specific use cases. Um, or maybe it's more of like a

cases. Um, or maybe it's more of like a a construction rig with a crane or something like that where you're I have no idea if you're intending to do essentially some more modular or someone else is doing like modular kitups or something like that. Um

but um can you can you speak a little bit like not so much to the camper because I I think it just doesn't necessarily fit but where you guys see like the trucking going because obviously it's the lifebird of the country.

Yeah. I mean trucking is an incredibly diverse industry. It really is. And I

diverse industry. It really is. And I

know that sometimes folks might drive down the road and think all trucks are the same. They're not. And they've all

the same. They're not. And they've all been customized and tweaked for that application because it's a tool to do a job. And so what we're trying to do is

job. And so what we're trying to do is provide the most flexible platform and ideally this truck and its standard range counterpart can serve the majority of needs. Now over time, look, there's a

of needs. Now over time, look, there's a lot of interest here at Tesla of electrifying everything. We want to

electrifying everything. We want to build every science project you can think of. We'll start with this and a

think of. We'll start with this and a big part of it also is the charging network, right? So a sleeper version

network, right? So a sleeper version would not be something that would be very viable right out of the gate because there's no over the road charging network to support it. So what

we're targeting is the dayc cab operations, the local hall, the regional hall, and as we build out the charging network to support it, there's a natural segue at some point to potentially look at a sleeper.

Yeah.

And then the other question that always comes up is autonomy. And that's really the other thing this platform is designed for. And by having that

designed for. And by having that flexibility and that forethought of okay, all of the parts in here are designed for autonomy. It's meant to be hardware ready where we have the necessary redundancies in steering and those other chassis critical

componentry. got the same hardware set

componentry. got the same hardware set in terms of camera type and compute that we do on the car side. So we can go through the same training and work to leverage the backbone of infrastructure there. So we can go in that direction

there. So we can go in that direction and at the same time we recognize we're going to have humans in the truck for a long time because we got to train that data and again there's so many diverse applications that people are going to be interfacing with these trucks for a

while and that's where other things might be a sleeper might be other vocations that we get into or work with partners for. We

don't know where the world's going to take us just yet, but we're excited for whatever it is.

Nice. So, one thing I thought was interesting, we were looking at the drive module, at least I think that was the highway one, just a single motor.

Um, so there is some commonality with the Cybert truck as far as drive modules. Yep. Obviously with the the

modules. Yep. Obviously with the the EPS, which I think this is the only I think truck ever that's got essentially we think so. Yeah.

Yeah. At least from what I know. Um,

like how how common or I guess how how similar is the front rack um to a Cyber Truck? So the it's common in that the

Truck? So the it's common in that the the motors are the same architecture. We

did beef them up. They need some more capability, but the same theory in terms of the way they're made. They're uh low voltage and comm support and redundancy.

And there's two of them and they're both independent and redundant.

That's the same as Cybertruck. And so

that's where again we've taken the learnings in the example, but we've modified it to fit this application. So

would you say like um and I'm not trying to overly pry but like boards things that are high investment like your boards and everything you need for redundancy relatively similar and then essentially obviously size whatever you

need for the specs like you know the the drive gears things that nature has been scaled to this vehicle but realistically the high investment portion is is very similar like you get a lot of carryover

I mean we really it depends it's it's a super case by case basis the abil the power electronics that are on board to support the low voltage system. Those

are carryover and those are, you know, from Cybertruck. There's small bomb

from Cybertruck. There's small bomb changes and maybe some programming changes to make it work for semi, but on the whole they're like virtually identical and we try to take that

approach and make as few modifications that we can, but then some components are direct carryover and where we don't have to change anything and we really try to just look at whatever the optimum is. And then there are times where we

is. And then there are times where we look and we try to square it every which way we can and we say, "Nope, we've got to do it on our And so the motor is an example. The stator is a modified

example. The stator is a modified Cybertruck one uh to just mechanically interface with the drive axle. So the

the bulk of it carryover in the same it's made on the same line.

Okay. Yeah. So all that stuff and then uh drive inverter very similar. Um and

then the rotor is unique to semi. So the

rotor is different and we have the rotor line here and it's just because the duty cycle of the truck required a new rotor.

Okay, that's all it was. And so we did take an example of the the stater that had line capacity, big investment that we wanted to leverage there. It doesn't

make any compromises. This truck is more efficient, provides more torque than our previous version. So it's not like we're

previous version. So it's not like we're looking at the shelf and saying, "How do we uh get by with doing something and cut away?" No, we're trying to optimize

cut away?" No, we're trying to optimize at every phase, but sometimes we can optimize without having to then go make big changes and investments. So I mean it's one of the things that I think we love doing and we'll I'll try to wrap

this up is just the iteration and we do it relatively chiefly when we kind of model we baseline redesign it and then kind of show capital investment all that stuff and it sounds like you kind of went through a lot of iterations with this cuz like looking at the drive

module when you see a lot of the other vehicles the other kind of over electric over the road trucks it looks like essentially it could be like a daimler axle or something like that where it's got a you know drive unit just been

essentially bolted to but nothing about this aside from essentially where you get into your your air brakes and outward to your more conventional braking components. I mean, it looks

braking components. I mean, it looks completely unique, you know, and tailored to this application.

It is, and that's where it it all starts from like the first component and a number of things that, you know, have you wouldn't think of a knock-on effect.

They do, and they're all related, but that's where Tesla's tight vertical integration, uh, really all matter. And

so, um, let's take the drive axle, okay, part of how we design and have to package everything relates to the frame rail spacing. And the frame rail spacing

rail spacing. And the frame rail spacing is mostly actually set by the battery and the way that has to quantize. And so

you end up in the way that we had to quantize battery cells and modules resulting in packaging for the drive unit. And but at the same time making

unit. And but at the same time making sure that that stator that we're borrowing from the Cybertruck line still fits not just statically but once that drive axle is moving around and sweeping through all the various positions that

nothing clashes and conflicts. But the

beauty of Tesla is that all those teams work together. So we know from the

work together. So we know from the manufacturing engineering team, yeah, you can use that stator. These are the modifications that we can make. We know

from the battery team working closely with the structures team on the frame rail sizing and spacing then working closely with the chassis and the drive axle team like all of them work together to make sure that those packaging then

all checked and validated by integration are all going to work and yeah it ends up being something that is bespoke but it's the only way to make something that is truly tailored where you get the lowest cost to build and most efficient because otherwise you're just making a

lot of compromises along the way because you're having to adapt to something that's off the shelf from a supplier that isn't optimized for you.

Yeah. And like every every company does that, but in practice I feel like they everyone struggles. That's honestly what

everyone struggles. That's honestly what I'm most excited to see besides the line itself. It just

itself. It just we struggle too. Don't get me wrong.

It's a lot of work.

Anywhere there's people, they're struggling. Trust me. Um, but that's

struggling. Trust me. Um, but that's what I'm most excited to see from the line perspective of just where you guys have taken where what silos have maybe been maintained in comparison to other trucks and then where they have been

broken down and melded in and just where between functional areas we see maybe because in general we just see it's a lot higher with Tesla than most folks.

Yeah. But at the same time, I have a huge amount of respect for the trucks that are on the road that are doing work day in day out. And they make I mean they make good trucks. They're reliable.

There's a reason they become the backbone of moving freight in this country. And we do try to learn from

country. And we do try to learn from them. And we really look at and say why

them. And we really look at and say why did they do it that way? And sometimes

it's like that makes a lot of sense. We

try to emulate it. And other times when you're particularly with a groundup design for electric platform, it's just not possible. Like the two are so

not possible. Like the two are so fundamentally different between internal combustion and electric that we have to go bespoke. But if there's lessons to be

go bespoke. But if there's lessons to be learned, we absolutely will borrow anytime.

Okay. Awesome. Listen, Dan, thank you so much. I know your time's precious here,

much. I know your time's precious here, so we really appreciate it and we're super excited to see, you know, the rest of it.

Happy to have you here and show it off.

All right. Thank you. I appreciate it.

Take care.