0:00
/
0:00
AAH #701 - No More Proving Grounds? The New Way to Develop Cars
AAH #701 - No More Proving Grounds? The New Way to Develop Cars
0:00
0:00
0:00
0:00
About this episode
The discussion centers around the evolution of vehicle development, highlighting the shift from traditional proving grounds to advanced technologies like ZF's Cubics system, which allows real-time adjustments to vehicle dynamics via software. Guests Oliver Spies and Lindsay Brook delve into the challenges posed by Chinese automakers' cost advantages and the steel industry's response to aluminum giga castings. They explore how these innovations can streamline development processes, reduce costs, and enhance vehicle performance, while also touching on the implications for safety and integration with existing systems.
Topics:
vehicle development
cost reduction
chassis technology
aluminum castings
steel industry response
software defined vehicles
real-time adjustments
Chinese automakers
safety systems
integration challenges
TOPICS: Vehicle Software PANEL: Oliver Spiess, ZF Group; Lindsay Brooke, Freelance; Gary Vasilash, shinymetalboxes.net; John McElroy, Autoline.tv
Filter:
|
Technical
Too Afraid to Ask
Car
Mercury Villager
"...wo years ago today, the second generation Mercury Villager went out of production. Now this is the part I di..."
Car
Nissan Quest
"...'t where you're going. Nissan, Nissan, the Nissan Quest, right, and they were building those in Avon Lake..."
Car
Chevrolet Malibu
"...nd you guys saw it with kind of tail end of Chevy Malibu, you know, where they were making really lightwei..."
Select text to request an explanation
Out online after hours is brought to you by bridge Stone Tires Solutions for your Journey. Hey Gary, what a show we've got today? We do?
He's gonna be an interesting one. Speak of interesting things, I found a
fascinating fun fact when I was looking for something that happened today, and I'd completely forgotten about this. And because you guys are smarter than I am,
I'm sure you guys remember this. But so twenty two years ago today,
the second generation Mercury Villager went out of production. Now this is the part
I didn't remember. I don't where you're going. Nissan, Nissan, the
Nissan Quest, right, and they were building those in Avon Lake, Ohio, and I'd completely forgotten bought them for a built them for a Nissan.
It was just a rebadge to Mercury Village. Yeah, And so whatever happened
to the relationship between Ford and Nissan, it didn't work. I think the
Quest never sold very well person neither did the Villager, and that thing ran its course and they both decided, you know, we really don't want to do this. You know, Nissan's got its own pride, it doesn't want
to rely on Ford, and Ford is like you know, we're diluting the value of our own Mercury Villager by having this Nissan thing that's identical with a different badge on it. And many Evans kept getting bigger, and I think
those platforms stayed fairly small, right And for those who are listening, that is the voice of our colleague Lindsay Brook. Yeah, who's joined us on
the show here today. We should tell them we've got Oliver Spies from ZEF
as well. Oliver, what's your title there? My title is on It's
called the heads of Technical Key account Management. So I'm leading globally or technical
sales departments. Okay, cool. So we've got a couple of themes here
going on the show. We're going to be talking with Oliver about how to
take cost out of cars. I mean, one of the things that this
whole industry he has got its tear on fire on right now. Chinese automakers
and suppliers of a thirty percent thirty percent cost advantage and so all the I can tell you here in Detroit, everybody scrambling right now, how do you take thirty percent cost out? So we'll be talking to Oliver about that and
then later with Lindsey. Although this will be sort of a combo conversation.
How is the steel industry responding to all these aluminum giga castings and the steel is not sitting on its hands. So we got some really good topics that
we're going to get into here. But Oliver, you know, won't we
first wanted to have you on the show. I'm very interested in this technology
that zenef has got called cubics, where you can literally, with an engineer sitting in the front passenger seat with a laptop, adjust the brakes, adjust the suspension, adjust the steering, adjust traction. So you know, instead
of the old way of doing things, you go to the rooving ground.
You got boxes and boxes of screen of springs. You've got boxes crates full
of different shock absorbers, You've got different sway bars, you've got different bushings, and you start driving them and combining them, and I mean it takes weeks, if not months to find the perfect setup. Right now you can
do it with a laptop. You can. But let but let me first
say a couple of things. What you said about the Chinese, right,
and it is not just that that Detroit is scrambling or even the West coast.
It's everybody in Europe as well, and even more so because in Europe rightwhere everybody outside of China is came back. I just was in Korea two
weeks ago and I spoke to Collex Are like from for example, from GM and all that same thing. So it's not just for us in Detroit.
It's a it's a scenario which we all have to deal with. Right.
Secondly, when I talk about Cubics and how to do that, just to give them my background as well as I don't just come from the supply, but I worked twelve years at the OEM as well, and I was responsible for that, right, So I spend a lot of time at the proving rounds, driving the cars, seeing how they build the new shop, obs overs and putting them in and driving again and doing it the whole day and doing it for months. Right. So the idea was born really many many
years ago, six seven years ago, back in Europe actually, where we said, you know, we have we have that development and Chussi which you just describe, which takes a long long time and it's always late. It's
not that pointing at the Chussi engineers, but in the program timing, it's always late. And now then over the last let's say fifteen years or so,
as is coming right more and more and more. And also that is
an area which is very test incentives, so you or a lot of testing you need to do, and it's always late as well. And now you
change one thing on one side, it potentially influences the other sides. Right.
So the idea was born from that to basically take that change, if you change the chassis part, or if you change an aid US part, take that out of the equation that you don't need to revalidate your whole car again, right and run through the loops again and again and again. So
the idea was to create an arbitration that's called it software between the two right and to make sure that you can deal with all of the actuators in the car on the chessa side at once right and not just like like you explain, not just do the suspension here and the breaking there and the steering there.
So that was really the the idea. And sometimes I use an analogy
basically, like you buy a computer right, you buy an Apple, you buy a Dell, you buy whatever you buy right, and you have a printer again can come from whoever. But when the computer says I want to
print and A, the printer prints and A. It doesn't matter from which
company they come. And that is an important message as well. So we
knew from the beginning when we started that development it needs to be an open an open system, right, So we cannot say you can have cubics from us and save time and money for your development, make it easier, make new functions and features, et cetera. But we we can deal with whoever
is a supplier off the actual actuators as well. So it's an open it's
an open book, right. You don't need to buy everything from that ft
basically to make that happen. Right, So those were all the let's say,
the starting points to put that all, to put that all together.
So but to be able to control all these things with the laptop in the front seat, don't you need a software defined vehicle? So yes, it
would be ideal to have something, but we actually started without that. So
the idea was was really to take that out. But now with a software
defined vehicle you even have more possibilities. Right, So, now with a
software defined vehicle, you can imagine that writing functions and features on top what I would call like the cream on the ice cream, basically just just on top of it is much much easier, right because it's all it's all service oriented and and it's much easier to organize. And again, we can do
it for the customer. The customer can do it themselves. Can even be
a third party. Right. So if we say there is a supplier who
does one part of the actuators and there a significant, a brilliant function they want to bring in and the OEM one set, we can do that as well. That was in the old days that was not possible. So so
I mean, before we get too far into the digits in the cloud and so on, Okay, if I'm thinking about a shock absorber or thinking about springs, okay, I'm thinking about mechanical devices basically it could be hydraulic or just you know, you're talking about making adjustments. Okay, So how do
you make an adjustment to a mechanical spring? How do you make an adjustment
to a hydraulic shock absorber? I mean, how does this happen? So
nowadays in the cars and the base cars, you still have that, right, you still need to work with your shock absorbs. You change the John
spump, the holds, the Jones spuns, et cetera. Right, but
many of the let's say more premium vehicles have electronically controlled actuators now, right, So your break is electronics. In a long time, you're from steering
the EPs. If we go one step further, we can talk about by
wire systems as well. On both of them suspension. Now, you find
a lot of what we call within our portfolio CDC, like continuous damping control, which means you have valves there which are electronically controlled. So that's you
don't change your blade anymore in your in your but you just do it.
Viare software, right, and so for example, the first car we launched was that in twenty two. The lot of Electra has for sure front steering
and breaking electronically. It has the continuous damping control, so suspension control.
It has electronically controlled swave bars, it has rear steering, and it has even an active spoiler. And I believe air suspension in some of the in
some of the variants. Yes, Well, it gets to Gary's question,
because that's a really good question. You change the spring rates if you've got
air suspension you're not going to you know, electronically control, you know, a coil of steel, right. But everything else, as Oliver is saying,
is you have to set basis right, the basis mechanically and everything is still set right. But then the tuning of that beyond that using all the
electronics, the electronically controlled areas of the chassis, that is done by Cubic set. And so you're saying that entry level vehicles are basically still done the
good old way if they don't have anything electronically controlled, they can still be done. But the advantage of Cubics, for example, is that you have
that range of vehicles. Right, So you have one platform where you have
one vehicle where steering and braking is always electronically controlled, but then you don't have an option for your electronically controlled suspension, for example, but one of the premium variants of your platform has it. Cubics knows that. So Cubics
will analyze the car and can run in the base car where you just have let's say front steering and your brakes, but it runs in the in the upscale one as well, where you have electronically controlled air suspension and CDC in there, et cetera. So it knows that. So for examp sample,
when you have an an AS system telling you there is a curve coming, the curve has fifteen meters radios and you're doing fifty miles an hour, and here's what I assume I send to the steering. Here's maybe I need to
break a little bit. Here's what the subpension that you're supposed to do.
Now you realize that one part of your actuator doesn't work anymore, it's broken for some reason. Cubics can adjust for that as well. So Cubics will
take that into account and can compensate for that and for example, tell the car to break or to slow down. So on the lotos we have control
over it's our electric drives as well in the car, so we have control over those as well, for example. Well, boy, you just said
something interesting. I thought Cubics was just a development tool for engineers. So
are you saying it resides in it? It resides in the car. So
it's really it's a it's a piece of software which can reside anywhere in the car. And that's and back to your question before about the STV area.
So today was the centralized architectures and the big let's say domain COMPUTERSAR, et cetera. It's hardware agnostics. So it can sit in a central computer.
Let's take a Tesla for example, it could sit on their own high performance compute. They're basically designed. It can sit in any of the x ravators
right, So in the LOTOS for example, it sits in our break module.
It's just in the break but it controls everything else. We have done
testing where we basically put it into our rear steering yeah ECU. So it
doesn't really matter where it sits. It's really just a piece of software which
controls the entire car. I was for getting to the initial subject of cost
on a vehicle program. Let's say two hundred thousand units, multiple models and
iterations off that. What sort of savings could the autoline audience who's involved with
this, what kind of savings could they realize over X number of months or years with this? So that is a question we obviously got from our customers
as well. And so and if you tell them a number, if you
say, hey, it's this many million you can save during the development there first you look at you and say, oh, that comes from a supplier, they may not trust you. So what we actually did is we actually
did I don't know then I worked on both sides of dependence, so I do understand. But we did a study, right, So we really we
went out and hired a company to actually do a study in that and we assumed a very high end vehicle with a lot of actuators in there, and we came out with a number and I don't want to say the exact number, but a number which is most of twenty million. Now an OEM can
say, well, I never spent that much money on my chrusity development.
That that could be right because you may not have all what we assumed there.
Scale it down right, So just scale it to what you maximum pay and we can tell you how much you can blacity save on that. But
that was in that example, which was done by a third party. Is
the number to save was norse of twenty million dollars. So this is required.
One of Jan's favorite topics zone architecture within the technology the vehicle. No,
it does not really so it really it can. As so the implementation
we did in twenty two, well, we actually when we did the development, we did it on let's call it normal cars, right, so they had just a can A network in there and get distributed the UCUS the LOTOS is still exactly the same. The next generation of vehicles, when the electrical
architectures in the next couple of years come out, was more and more domain controllers on centralized architectures. They will work slightly different. We can do both.
We can do the old we call it signal based, so where you really connect to everything through a CANFT network for example, or we can do a full high end eason it based domain controllers as well and service. We
talked about car savings, but seems to be at least in the development phase.
It's the time savings that people that companies are going to go after.
That is part of the costs as well. Right, So whenever we times
money, yeah absolutely, but you know it's how do you get to market faster with the latest technology. So in China, like you said, in
China, the cycle times for a car is down to eighteen sometimes twelve months.
So the customer comes to us and says, our one isn't this part but I needed in twelve months to start a production. And a few years
ago we stood there and said how is that possible? Yeah? Ourselves?
Right, So how is that? How can we do that? And nowadays
we are able to actually compete with our low called local Chinese competition basically do and we have delivered programs between twelve and eighteen months, and Cubic's for chassis is one is one part of it, right. But the time, the
time reduction has really seen more at the OEM than that our side, right right, right, But if I'm one of those guys who's got to go to the proving grounds with crates and bushings and shocks and all that, what kind of time could I see developing a chassis, chassis, steering, breaking traction, all that? Again, if you're without saying names again, take
some of the traditional oms in Germany or in or let's say Europe or here in the US. You doing a programmers in three four years, the Chinese
do it in eighteen months. Right? But would this facilitate getting closer to
absolutely? Yes? And that's how we try to market it as well,
to say, guys, so it's important, right, So we are not trying to tell the industry it's our DNA. It's now the car drives like
a headf car. The DNA is still from the customer because they do the
collaboration they do is the tuning. They all do that, but using that
tool right to basically to do that, and now that you bring everybody together doing it in parallel rather than do this first. As as to do this,
that's when you can shrink the time. Will this bring chassis development farther
up the development stream, more towards the front end and the back end, or will it make chassis engineers say, now that we have cubics, we can put this off until the eleventh and a half hour. No, I
hope they won't. I hope they won't use that. So there are two
different things we need to talk about. Cubics is basically you need to split
it in two different times. One is virtual developments and one is real hardware
development. So Cubics is available for both. Because your early development still you
want to push that towards the virtual phase as well, right, you want to do that virtually. You don't want to create prototypes early on in anymore.
Another cost and time reduction, right which you need to fulfill that eighteen months cycle time, and then when Cubics is in the car, you're quick again. And I hope not that the chaszease guys will say no, I
can wait for three months and then I'm doing it. That is not a
solution. So I want to go back to what you were saying earlier about
Okay, somebody's driving the car and a curse coming up ahead in Quebex somehow knows the curves there a how does he know it? And then b okay,
Cubic says, Okay, it's going X miles per hour and ideally it should be going X. My is something mouths prior to take that curve,
but Lindsey's behind the wheel. He likes to hammer it through curves. Is
Cubics going to prevent him from being able to do what he wants to do?
You could potentially program it that way. But let me back up first.
So the assumption is there's an a system in the car, right, So not Kubics knows what the road ahead is, but the aid of system does. Right, So the AIA system will read here's what's happening now is
the trajectory control. So you need to make sure that your chassis follows that
trajectory control, that which is planned by the system. That calculation can be
done by cubics, so they can basically say I have only I have a fold, but the guy is still driving towards the dealership, and I have only fifty percent of my front steering available dual bon motor or something. I
lose fifty percent, so I can't steer any more act really nicely at that end that angle, so I need to slow do on the car, or I can use the brakes to steer. So the benefit the development timing,
the cost reduction is for me still was the number one idea of doing that.
But those new functions and features which you can now do cross actuator, that opens up a whole new world. But that is then better in the
STV type of vehicle because it's easier to control and those new functions and features again that f can support on that or the OEM will always say I can do that myself. Absolutely, you can do that using our tool, you
can do it even better. Oliver. The concept is that moving very quickly
in the last ten years on aid ASS technology, I mean really kind of this was a wellspring within the company. Did the concept come more from the
aid ASS getting together with the conventional vehicle hardware guys or did it come from the chassis guys at the time, because I was involved in that early team who actually came up with the idea. It more came from the Chassia side,
especially because of the what I said at the beginning, begin to be late during the developments, always doing changes and all. There is that eights
system which does changes to us. We need to comprehend for that. So
it was more driven from the Chassa side. You talked about simulation earlier,
moving it all to the virtual world. That's where the race is right now,
right is absolutely how do you simulate as much as possible using digital twins?
Can you take cubics and work with a digital twin and do the chassis and Look, everybody wants to do the simulation, as you just mentioned a moment ago, get rid of prototypes or cut way back on them. There's
some talk of maybe even going to zero prototypes. We'll see, we'll see,
but there's talk about that. What about plugging cubics into a digital twin
and doing all this development without even going to the proving ground. Yep,
that's how the development starts. So we do that on Hillsale. So why
do you still have to go to the proving ground the correlation is still not accurate enough, or when you get to the hardware, there's variation. I
come originally from the crash side, from passive safety, and we did a lot over the last thirty years in terms of putting it virtual. Right,
I'm not crushing cars anymore. You still crash a car at the end of
the day, so you still drive the car. You still I mean,
the government still requires it, right. I don't think they'll accept simulate where
there are discussions in the industry to try too, but not yet allowed.
But so what I'm trying to lead to is is really your final validation.
Right, there are certain things you can't simulate yet. What we can do
is what our aim first is when you take the development, you have what we call concept development. You have design validation and process and production validation.
So DVPV, the CV and the DV we're trying to virtualize. And whereas
for some components we're at one hundred percent, we can do that. We
can skip it. Right, if we review it was a customer, we
can skip it. A PV will still run mostly in hardware. Maybe we're
too conservative in the in the industry. I don't know, but that is
how the aim at the moment is right to take out that time at the early development stage, but still do the validation mostly in hardware supported by Hill and Sill as well. Right, But hopefully you're doing that validation with the
stuff that's going to go to production, correct, so you don't need to then do testing and re verification revalidation that would then drive the timing out again.
Correct. So what are you guys doing beyond CUBS. I mean that
does a lot of things, and this obviously is just one small part of it. So the one thing is really too when you know our portfolio,
right, we have we're a very large chassis company. We have electrified power
training, we have vetas, and we have integrated passive safety. A lot
of those fields are let's say connected somehow, especially when you think about chassy and power trying and striveline right, So those things, there are things in the works in terms of how to combine those things and how to make that even more efficient, more cost effective for the future. And then obviously how
what is the interface to the aid system. And then there is an a
link as well between aid US and passive safety. That's what we call it
integrated safety. So there is there is a link there as well, from
coming from the active safety side towards the passive safety side as well. So
there is a lot of that ongoing to see how the portfolio with the synergy is there. But then the big thing what's happening in the industry is what
you like the software defined vehicle right in terms of how do you how do you construct your software in the future, and how do you do what other people called digitalization. How do you make money with the data you have on
the car? That is a big topic as well. You know, I
just went to the to the artitect a couple of weeks ago, and when I hear the executives from the OEMs talking how much money they want to make was digitalization. We will see how that works. But but again there is
a lot of like I call predictive maintenance for example, the holy grail of digitalization if you make it happen. We're in some areas very close to to
have some solutions that is that you can really sell and it would benefit the end customer who drives a car. We talked about the cost advantage that Chinese
automakers have when they acquire cubics. What happened do they have even more of
a cost advantage? Well, it started in China, right, was the
first one to use it? Is that right? We have China, Yeah,
and we have another Lotos, and we have another couple of companies in China where we work with as well, and they have brought cubics in your into production. I think I said it before, right when the you need
to compete with an was the way the Chinese work. I worked a lot
with one company there one one am because I worked for them, and they always said, look, you cannot compete with me. I work from nine
to nine and six days a week, right, And that's how the hole thing starts. Because it's not that the Chinese are cheap anymore. The cars
are not cheap. I mean, I hope you have seen the cars.
There are no low cost cars over there. The latest generation of cars can
compete with the cars here and then Europe easily. And that's why they buy
a lot of high end stuff. So it's really it's not because they are
low costs, you know they're not, right, It's about how they work, what the attitude is, right, If somebody says there, I want to change my software. The whole company turns around and in three days later,
or maybe in twenty four hours, you're having new software. Some of
the traditional OEMs, some of the traditional suppliers need three weeks to make a decision if we even want to change software. So it's like it's part of
the attitude as well, right, So, and you really you have to work there. You have to interact with those guys over there to understand how
it works. I can tell you some more stories you want, but it's
really like just one example, we did a car fifteen years ago. I
did a car for them for the Chinese, or did the Chinese market, and my going in position and I was responsible for the vehicle performance. My
going in position was or so I make it lower cost, I make it worse, the NBA should go up and everything. It was exactly the opposite.
What do you want the car to be better? You want the car
to be better here as well. Okay, I totally misunderstood your originally indication.
So it's yeah, yeah, it doesn't come some people saying, oh, it's all cheap and what they do over No, it's not. So
it's really they know how to engineer. They don't cut corners. They know
how to do it. They just have a very different attitude, boundary conditions
and they go through it and we need to learn to adapt to that.
Which we did was our teams, our teams in ZF in Shanghai, and we have a couple of engineering centers in China. They know how to do
it now. They know how to turn around a development program between twelve and
eighteen months and it's amazing to see how they can do that. It's a
well well oiled machine to make that happen. That's right. I want to
go back to cubics a minute. We've been talking about chassis, steering,
breaking and all that. Does it go beyond that? I mean, what
other parts of the car could you use cubics to work on? So at
the moment, we because we have like the six degrees of freedom, we always talk about chassis, right, so in all directions the sixty degrees of freedom, we have all parts within our portfolio. So that's how we started
off. Obviously on the lotos, we already branched into the drive train as
well, right, So at the end of the day, we can control the drive train as well. The Lotos already has an active spoiler for example.
We don't have anything to do with that. We never did one of
those, but Cubics controls it. So again, it's about combining and finding
synergies between the actuators to make them work together. And that's why I'm saying
we are not dictating on what you need to do that, right, We just give you the ability to do all that. We can do it for
you as well. But sometimes the customer feels like we take over. No
we don't. We can, we can offer that, but you still want
to tune your car. You still want to drive and have a Cadillac drive
like a catileag or a Lotus like drive like a Lotos. You don't want
to drive them both the same like as the f car. That's right,
that doesn't exist. Do you see a potential integration of Cubics within the hardware
in the loop and software in the loop industries? That's what we already had
do today, right, So coming back to the question with those vendors, I can't say much to those vendors. I mean, there are multiple vendors
out there. We obviously we use a few of them. We don't just
use one of them, but there are I don't want to say any of your names, but we are integrated with some of them, not just with one. So we need to make sure that we stay flexible there as well,
because not every OEM has the same time tool chain and that's maybe a topic which is very difficult for the STV. Right, So the STV requires
a perfectly aligned toolchain throughout the development for software especially right, and that will never be the same for everybody, right, So a GM will work it differently than than MSU T spends on a BMW's and a Chinese one, so which means we need to be flexible. Right, So we need to be
able to work with multiple heal intil the facilities basically or providers. So you're
dealing with safety critical systems on a regular basis here. So does Cubics take
over the intelligence, say for a breaking system, does it take it over for a steering system, or does that responsibility still reside with those systems?
Very good question. So we when we talk about Cubics, we talk about
let's say three different areas. One we'd say is Cubic's framework that just allows
you to implement the tool basically right. Then set the boundary conditions and make
sure that everything is in its order, where the actuators are, where the arbitration is, et cetera. With that comes what we call cubic safety.
So what you were just saying, so we provide that as well as an envelope that everything stays within their limits basically right. So, and it's important
that whatever you change because in the future when you want to do an OTA update, so you don't want to revalidate the whole thing. So it's the
safety corridors. The cubic safety stays. It doesn't matter what part you update,
so that will always govern where you are. And when we normally do
first evaluations with customers, with potential customers, that's always one of the big questions, So how do you guarantee that my car is not going off the road? This breaking doesn't go into that direction, And we always show them
first how the system works with cubic safe and how the system could work if you don't have that part of it. So yes, it will control that
we are not we are. Basically we have interfaces to the breaks, right,
so we tell the break on what to do, how to do the valves, how to do the motor on the breaks. If it is a
different supplier, that's the supplier are still doing that. We just tell the
breaks what to do, so we do not go to the level of actuating their motor or actuating their valves. Obviously we do that on our parts,
right if it is from that f but not if it is from a different company. We just tell the components on what to do, not how to
do it, not how to do it. In the system, most of
the components work the same way, right. There are in some areas there
are some differences. But a break system works like a break system. You're
the maestro of an orchestra, do the musicians still have to play the instruments correct? So again, that's why we can control an active spoiler, right,
because we understand together as a customer on when do we need it, when do we want it? When you're on a race strike, you drive
too fast, et cetera. You want that thing to come out. We
don't need to be We don't need to understand how the mechanics or the offset works. So does this save code development on the OEM side, Because some
of these subsystems are controlled by other other controllers current chassis control, powertrain controller, et cetera. I think the more we can bring into the control,
yes, it will save on the on the OEM side, and it will centralize us. Right, so you don't need to have really certain things sitting
somewhere. You have one centralized software which will control your entire chs and drive
train and whatever associated things which have an influence on your driving behavior. And
that could be z f cubics. That could be zetf cubics. Correct.
Hey, look, we got to take a quick break here. We're going
to be talking about more a programming note to the audience. The autoline crew
is taken next week off. There will not be an autoline after hours next
week. We will be back the week after that. We're going to be
back in a minute. Here. I'm dying to ask LENDSI about what steel
industry is doing here to compete against giga castings, aluminum gig castings. What's
the steel industry going to do? We're going to be back right after this.
When the elements are working against you, being confident in your grip on the road is what really matters. Breach through the lens of tires improved acceleration
in wet conditions. All right, we're back. So yeah, I mean
I set you up just before the break here, and uh, I mean you know what I mean, Yeah, yeah, well yeah, I mean everybody's going after castings. I mean, you know, I was watching what
Rivian's doing this morning. They had their Investors' day or what a capital markets
day, and they're talking about how many casks they were putting in NOD.
But you told me steel industries got, you know, some things it wants to talk about. It's not sitting on its hands, right, And there's
probably some backstory to be told here. And Gary and John have followed this
since the nineteen eighties in terms of steel fighting back against threats from other materials plastic, aluminum. We went through plastics with bits and pieces of component fy
fenders and so forth, and then some big volume programs from GM Saturn the whole Saturn landscape, the minivans, etc. And then BMW getting into it
with a lot of So steel was kind of on its heels then and really hadn't applied R and D at that point in the eighties. Well, they
got religion really quickly with the scare from plastics and really innovated on the R and D side, new alloys, new processes, and of course they had worked with the OEM since the beginning of time, so there was that relationship there on tooling and how to set plants up, how to do things a fish efficiently, welding, forming, et cetera. So they were kind of
on track with that, but they really charged back and then in the eighties you had PNGV and high mileage programs, and of course, of course in quotes, they were going to be aluminum because it was optimized one hundred miles per gallon cars. How are you going to do this? Super lightweight steal
innovated again and kind of headed off that pass more recently f one fifty lightning all aluminum. Is this going to change the landscape? Well it really hasn't.
It did it for it with sheet metal, but it hasn't underneath the vehicle. And Tesla comes along, and Tesla initially composites and aluminum, and
then all of a sudden they go to steel because they're building lower priced cars, so they kind of gain gain a proficiency was steel attain area steel cars, but at the same time developing this giga casting, which I should have I should have brought those of us who built scale model cars in the sixties the basic subfloor of the vehicle, which as we all know, was injection molded, but it was the front cradle, the rear cradle, and the
underbody all in one piece of plastic. Well, if you think of that,
that's what gigacastings are. And I don't know whether Sean was able to
pull them up, but this is not entirely a new idea to cast big parts of the car, particularly the whole underside and chassis of the car, in aluminum. And there was a guy named Albert Grigoire, French designer.
He was a lined with the luminium industry who at the Paris Show and either thirty eight or thirty nine showed a small car weighed a little bit over nine hundred pounds and it was mostly an aluminum casting. And I got to believe
that when the Tesla guys set out to do patent investigations on gigacastings, they had to look at everything. And there's a lot of Grigoire stuff online that
can show you just kind of google that GRE E. G. O.
I R. E. Albert and you'll get just the wealth of stuff with
photos of this really to save the complexity of a ladder frame vehicle or a steel fabricated vehicle and really condense it into a few parts. And so Tesla
comes out with this and it's an explosion, as you said, and companies like Idra, which is a Chinese owned Italian high pressure die cast pressmaker, and some others is too. LG has a big press maker. All of
a sudden, the tooling explosion is happening around this Tesla announcement, and it seems very well aligned with battery electric skateboard type vehicles. As a matter of
fact, Brad Davey, who's VP of Urcello Mattau Major Global, one of the biggest global steelmakers, says, you know, gigacast aluminum high pressure die cast can be quote an attractive solution. So he admits to that to reduce
floor space in the plant, fewer welding robots and does design wise lend itself to a basic skateboard architecture. Currently now we're seeing vehicles with die cast front
cradles, die cast rear and the battery boxes in the middle, and it's joined in some way and generally it's aluminum extrusions. Okay, save that thought.
However, the steel guys this was not news to the steel guys when Tesla made this announcement. They knew this was coming, and they really were
looking at kind of the legacy of what happens when this happens. And this
happens, we're going to have, you know, three components that are all die cast in a vehicle. They kind of saw this coming. They talked
to the tool manufacturers. They knew it was coming, they say, and
we can talk about this more difficult to make mid cycle changes and body changes.
Much more has to be changed. The tooling is very I mean if
you use Giga cas, if you use Giga casting, the tooling is very expensive. The machines are enormous, the steel guys say. Some Chinese OEMs
have already purchased Giga casting HPDC equipment between twelve and thirteen thousand metric tons.
These are even bigger than the ones that SpaceX has. Musk has bought these
for Austin Berlin and for SpaceX as well. So these are big machines expensive.
You could surmise that once the whole industry transfers over to this. They've
already spent the money, but getting to that point is going to be a big buy in terms of the equipment. But they acknowledge that this is a
threat to them. Disadvantages. Part costs are individual, part costs are higher,
lower yields, higher, part defect because these molds are enormous, and filling these molds is CFD mold software challenge, and any little bit of porosity in these big molds can be a problem. It could be a completely failed
part. And you're not just talking about a suspension control arm. You're talking
about an entire underbody of a vehicle would have to be scrapped. So that's
a big You know that that's a big worry. And to ensure that these
that the mold flow is high enough, some of these castings are being designed with wall thicknesses that are much fatter than a typical than a typical mold a smaller part mold, large die casting is very difficult to repair. You wreck
that car with an entire underbody of aluminum casting, it's going to be expensive insurance costs going up. And the steel guys also acknowledge that battery enclosures die
cast are aluminum dominates that that market right now, but they're going after that.
And they said they know one OEM that is completely going to transform over to to steal battery enclosures versus aluminum right now, just because they want to make them cheaper, because they want to make them cheaper. Now, So
is everything you said so far because they want to make them cheaper. No,
And I'll get to that. I'll get to that. So so ironically,
one of the steel engineers or one of the steel engineers is talking about needing to design load pads just to protect the castings in the underbody of the vehicle, not just to protect the documents, but to protect you know, the battery, which is going to be in kind of an integral die cast aluminum enclosure. So pros and cons. Okay, steel fights back. How
design again, innovation. There's like thirty five hundred steel alloys that are in
the market right now, and most of them have come about in the last twenty years. So they're we'rerarnning a lot of R and D into into new
alloys, but also new processes and Gary you're familiar with this, you know, laser welding is moving on press hardened steels are moving on. Martin Cidic
steels are moving on, still high cost, but volume is bringing them down because the industry is moving them into things like rockers and sills, and so volume of those new materials is coming up again. They're going after battery enclosures,
designed for manufacturability and designed for part integration. So they're seeing the advantages
of these big die castings and saying, with these steel alloys and with these processes, we might not be able to get four parts into one, but we can get four parts into two and moving it that way. And they're
very confident of this. The main reason is steel has to respond. Steel
is about fifty percent average the massive an average vehicle right now. They can't
afford to lose that at all, so that's really the driver behind this.
Obviously, these high tech steels cost more than carbon steels, so they lose that advantage gary a little bit on the material. But they think with the
matrix that steel has in the industry right now, from transfer lines, welding lines, stamping, tailor welded blanks which steel companies are buying into our Cello Mattal has their own tailor welded unit they're bringing all this together into kind of like an engine that will they feel we'll compete with with gigacastings. So when
we look at if we looked at a Tesla giga casting, we'd see this basically a solid three dimensional thing, right, right, are the steel guys talking about like a I'll call it two D for lack of better term.
They're not talking about making a casting. They're talking about elements that would then
be assembled so it'd be stamped right then welded or bolted or bolted and welded or right and looking at individual parts of that and saying, right now we have three And that's exactly one of the problems with these big guy castings.
If you want to add a couple of brackets to that, it's not just fabricated and welded a it's a new moment or it's part of the mold that goes into that, and that's expensive to change. So, yeah, they're
looking at they're looking at part Integration is a big part of steel, and you know, maybe you should have a steel guy on because they can tell you kind of the how they're going to get there with integrated parts with steel.
Because we think a part Steel assemblies as fabrications of sheet, maybe some cast parts, maybe some forged parts, a lot of welding, maybe some adhes a bonding, and that's still going to all be in play, but they feel it's going to be in play at a cost point that at least for the ramp up to BEVs that are going to be giga cast, they'll be there'll be a cost advantage. Well, I'd like to see that,
because you know, the fastest way to take cost out of an electric car is cut down the battery size, and the fastest way to do that is take weight out of the vehicle. I mean overall efficiency too. But if
you can take weight out and shave off you know, X part of the battery, you save so much money. So I buy what you're saying there.
Of what the steel people are saying, the peace cost for a giga casting is much higher, much much higher, no question, But you have to look at the total overall cost, you know, take it from a system standpoint, and I'd like to see what Steel comes up with. Well,
and that's what they're doing. And they showed some stuff. And another
thing about this conference, you guys know, is they always have an OEM who kind of lays out everything about a new vehicle and done in steel.
You know how they did this, and they had the chief engineer from the Equinox EV program. And you know, GM's playing it both ways. They're
doing not maybe giga castings, but like many giga castings on the high end Cadillac. It's going to come out next year. Uh, not entirely die
cast underbody though, I mean it's kind of a composite right now. But
they're also doing a lot of work in developing steel structures that will be for lower cost electric vehicles. And she said she's really bullish that this is going
to be a horse race because she's not giving everything to casting. She says
there's a place for them, but you know that's kind of where it stands.
Yeah, I think the castings are here to stay because so much partial part elimination and you know, all those stamp parts need die sts. I
mean a lot of expense, there, a lot of material handling, a lot of assembly time, blah blah blah blah blah. So my own guess
is castings are are are part of the game from here for well, there's no pot yeah, but to your point. Since the nineteen eighties, the
steel industry has been very innovative and aggress and I cannot wait to see how they respond to this competitive pressure. Right. Well, you'd have to look
at an entire vehicle structure and you know, have a graphic with callouts to say ten percent from here, fifteen percent from here, and how does it all add up one versus another? You know, because you got to buy
these machines. The machines are huge, I mean for cast for giga castings.
Yeah, well you know that's why Toyota is looking at doing castings.
They don't call them. I think they call them mega castings. Yeah.
Number one, they don't want to use the term that Tesla has already introduced, which everybody is used by the giga plant, giga batteries, you know.
But well Tesla certainly has, but you know, didn't Godzilla did get with Megatron, didn't fight with Gigatron. Right, But anyway that Tesla or
Toyota is looking at, we don't have all these big presses, but we do have these small I mean casting presses machines. So they're looking at how
do we do what kind of advantages can we get with going with castings with the equipment that we've already got, and they did, as you know, they started from scratch, whereas everybody else has the legacy of kind of evolving.
Tesla started from Tesa started from scratch, right right. I think it's
going to be an interesting horse race again because the steel industry has so much to lose. Think of fifty percent of the mass of an average vehicle going
down to twenty percent because of or even less than that. If if that
battery enclosure which is now extrusions, if that became a cast part of that you know, model car UNDERBIX. So I was just watching, as I
said earlier, the Rivian presentation today on their second generation battery pack. They're
going to not full aluminum extrusions, but the caps on each end of the pack are going to go to that And I don't remember the number off the top of my head, but yeah, they get rid of all these stampings right right. So I'll only ask you in your career dealing with companies of
different types, So giga castings or mega castings or what we want to call them, I mean, they represent a huge change in terms of the mindset of the engineers who are involved in developing these things. And as Lindsay pointed
out, these machines are very expensive. And these guys have a whole bunch
of stamping presses and welding robots and you know what I mean. So they've
got a lot of sunk costs sitting there. So how do you find the
receptivity of people within organizations to embrace big changes like he's talking about radical So it's basically it's typical tesla so so and I wasn't part of it, but I crashed cars for twenty years, so I know what a giga casting advantage over steel is and vice versa, right, and what the what the problems on ayes a side are If you need to have the ride and mindset on terms of engineers can do anything right. We can do anything right. Just
ask us a question and we will develop it right. And then you just
need to give them the boundary conditions. If you leave the bud conditions a
little bit open, you can think outside of the box. You come up
with a giga casting right because it has significant advantages right in certain areas.
Yes, you have to have this investment, like you were just discussing et cetera. You showed all the pros and cons right, But if I'm an
engineer, I would choose gear casting because it can it's so much better to optimize because then I go back to the to the to the to the steel side and say, you said it as well tailor way the blanks, right, Yeah, but it's not enough. I want more right because there's so
much inefficiency. And some of the old body in whites would basically have right.
They could be so much better, but I couldn't make it right.
I couldn't manufacture it the way. So the steel industry has to come up
with new ideas on how on how to do that right, on how to get some of the advantages. They won't be able to do everything right,
but some of the advantages the coming backs. Again, the attitude. The
attitude needs to be like you need to have an open mindset out of the books thinking. Don't say immediately when you run against the first roadblook saying okay
it's dead. Oh show the roadeblok. Show what the problem is. Weigile
solve that somehow if we really want to. And that's why I said to
typical Tesla. If Tesla runs against the rope block being a supplier or somebody
else that will just do it themselves, and I will do and they will.
And I should add too that the steel guys were saying it. And
if this is reported right before that Great Designs and Steel Conference, Chinese OEMs seem to be backing off from you know, from some of our Chinese OEMs are going forward. Tesla seems to be backing off from the giga casting push
a little bit, and that might be a distraction by must Musk. I'm
reading that differently. So yeah, I mean there had been reports that Tesla
was looking at doing the entire chassis as one. You know, yeah,
first they did the the what was it, the rear cradle, then the front cradle, then the battery box, connects, the connects and all that, and then they were looking at so the reports went looking at a total one. And now the word is they've backed off that. I'm reading that,
and I don't have any clue as to whether I'm right or not.
This unboxed assembly process that they're talking about where they're going to build the car and essentially one, two, three, four, five, six modules, Well, if you're going to have six entirely separate modules that only come together at the very end of the process, and Valla, you've got a car, then you don't want a full chassis. And so my read of it
is Tesla was going down this route of saying, hey, you know front cradle, rear cradle, middle cradle one craterle one. Then they start thinking
unboxed the process and they go, wait a minute, we don't want a full full chassis and not as one piece. So that's what I'm thinking,
is driving there think yeah, the still guys too are looking at why give them these front and rear cradles? And one example is control arms and you
guys saw it with kind of tail end of Chevy Malibu, you know, where they were making really lightweight control arms out of steel. It had holes
in them and so forth. They were light, but they were basically on
part in terms of mass with an aluminum casting. They're looking at the same
thing and taking that going after that business. Why give these guys that bi
So you're saying same performance, yeah, maybe even the same mass, but much lower costs or close enough mass. And I'm not sure with the relative
densities of aluminum steel that you'll ever get part to part, you know, as light as light as aluminum. But everything else that I mentioned, you
know, faster model changes, maybe some more manufacturing flexibility that it'll be for the customer, it'll be well, let's pick this one over this one, so not to be disrespectful to my fairest friends were busy making things that are magnetic. Yeah. So so basically what you're talking about, though, is
is that there has to be a whole different approach to designing these components.
Okay, And you know, the steel guys over the years used to say, well, you know, we've got this all, all the recipes are there. You don't need to change anything, you know, just just buy
this different steal and you'll be good. But now, based with this technological
challenge, big challenge, they're suddenly saying no, no, no, you've you've got to change everything, you know, whether it's an alloy where you know, whether it's a third generation steal or or drilling holes into something that you know you never did before, you change that. And so I just
wonder that if in automakers looking at that and saying, well, these steel guys have to they're telling me I have to do all these changes. Why
don't I just go to this other technology that these new cool OEMs are using instead of trying to stick with try. I'll mentioned working with OEMs from z
f's perspective, and there they're literally embedding materials scientists and design guys in with vehicle programs at the OEMs. I wonder if that's enough though. And the
reason I bring that up is remember in the nineties, the steel industry hired Porsche Engineering right and said, you work with us, the steel companies, and let's figure out how to completely redo a body in white. And they
hired Ricardo and they had a lot of these programs for light weighting, but they went outside of the car companies. Is what I'm getting at. Porscha
Engineering is actually it's part of Porsche, but as you know, it's Right Engineering. They do engineering for anybody who wants it right, and Porsche Engineering
came up with a body in white that was significantly less mass. I can't
remember where they were on cost. It used more high strength steel, but
they were able to take out all kinds of stuff. But it took the
steel industry going to Porsche Engineer, and then the legacies all looked at it.
And then went, oh my god, look what Porscha did. That's
so cool. We got to do that too. And I wonder if that's
what the steel industry has got to do with its war against the giga castings.
Is I work with somebody and because this is a show me industry, you can talk to your blue in the face about we could do this, but if you show them the hardware, then people become believers overnight. Well,
I picked up from these guys from davy at, Arcelo Matur and the other guys, mattal Sorry, this urgency, this kind of aggressiveness. And
every year at this conference there's a theme, you know, here's the challenge, here are the successes, and they lay that out well this year at the top of the pops was giga castings, was big aluminum castings. So
that was the perceived challenge laid out right at the beginning of this conference.
So that kind of tells you that's where their focuses. Yeah, the target
is on that right now. And then talking about you know, working on
battery enclosures, that market is like ninety nine percent aluminum extrusion, so they're looking to go after that and that that's currently the connector of this and then going after with design for integration with steel the other ends of the vehicle as well. No, I mean I can just tell you that it was an
urgent, urgent discussion from the top of the steel industry. We'll see if
it if they can play it out well. The like I said, I
can't answer some of these questions, but I mean it's it's it's going to be here speaking for the steel an interesting battle, you know, and this this was really you know, the three of us covering this back with it with steel versus plastics. This has the same kind of you know, we're
going to war. We're fixing bayonets on this one. Yeah. Well,
they also have the big challenge ahead of them in terms of green steel and in because I just wanted to bring that up so that they talk about sustainability, because there's a there might be a USPS that can develop, right because on both science, I mean a challenge. We are talking a lot about
green steel, about green recycled die cost material, both aluminum whatever else.
So well, that's right, that have bought a whole lot of green steel as every culture green steel futures. I mean that the steel isn't even there
yet. You guys are saying this is what we're going to use as we
because we have to use it and we need to use it to a certain volume to make sure that we get to the point where it becomes neutral.
Basically, so the industry has to change, right, So I wonder how that influences that discussion between a big die cost part or or making it out of steel. Well, they've got other things. I mean, they're investing
in small new nuclear reactor technology because the steel industries the mills are so energy intensive, and I think aluminum has that same problem. The manufacturing process is
newmanhum I believe is more energy intensive, much more energy intensive. So so
those industries in general. And then as we're discussing, trickling it down to
designing and developing vehicles as well. So interesting. That's why I keep saying,
this is the best time ever in the industry. Man. There's so
much stuff going on like never before, you know, and mentioning mentioning Porsche, you guys put transmissions in portions for people who are not interested in electric cars. You guys are still put not just in posches. We've put it
locally here in every you know, Ram truck. We're talking faster for farms
here, building, sport and berg. So that's for example, the big
dmw's as well. So yeah, so yeah, we have multiple customers,
but again multiple customers for transmissions, for hybrid transmissions, so plucking hybrids and the next generation electric drives as well. So we can handle the door at
all. We need to, we need we need to be flexible. Right,
So that could take another probably one of your shows, and just to talk about the EV slowdown. Oh what is that? Oh well, we've
been doing shows on that, right, who whole discussion. The industry needs
to be flexible? Is it? Is it? Is it flexible? Let
me say you need to be I think that answers the question. Yes,
But there again they're at there are some there are some things. When I
stand in front you know, of customer executives and I explain our transmission technology, is there? We were very flexible, right, We designed the part
which fits in any configuration. Right, because would you be able to sign
the paper saying in five years I built that many hybrids and that many EV's of the same platform. Yes, you can have a guess. Will the
numbers work? No? They won't so, which means there needs to be
solutions like that, right, so at least for the next decade or two, right to understand how the whole market go us. So this sort of
goes back to what John was talking about earlier though, I mean in terms of reducing costs, because if you're making these guesses of what the viumes would be in five years from now, and you know you're an OEM and you're capacitizing for that and you're selling a third of that, I mean there's a lot of wasted money there. Well, So having accurate assessments and or the
ability to change to something else then protect your capital. And that's a whole
another discussion, which is to supply bys has was the the OEMs in terms of you didn't deliver. Oh you want to many of them, which is
great, but there could also be a problem, right, So bose ways is a is a potential problem. And yet return on capital investment is very
important, especially and let's say the last five years of crisis crisis crisiss, that's right, never a dull moment. Hey, look we got to wrap
it up, Oliver, Thanks so much for coming on the show. Great
is the Great having you here too. Thanks. Look, Gerry will do
this again in another week. And by the way, for the audience here,
remember we're going to be off next week. But you don't get this
kind of programming anywhere else, so please, if you really like this, subscribe to where channel. You can make a donation on both YouTube and on
our website and thank you if you do so, I'll Do'tline after Hours is brought to you by bridge Stone Tires Solutions for your journey
Request an explanation for:
3 cars
3 cars featured
Request an Explanation
Heard something you'd like explained? We'll add it to this episode.
Sign in to request explanations for terms you heard.
Want to learn more?
Browse our glossary for plain-English explanations of automotive terms, jargon, and concepts.
Help improve this episode
See something that's not quite right? Our annotations are AI-generated and can sometimes miss the mark. Click the flag icon on any annotation to suggest a correction.
Report incorrect info
Suggest better explanations
Flag missing cars