- There's no doubt that aerodynamics have played
a huge part in the improved lap times that we've seen
here in the cars at World Time Attack.
There's a misconception though that developing
a proper aero package requires an F1 budget.
We're here with Andrew Brilliant from AMB Aero,
and we'll find out how developing an effective
aero package isn't outside the realms of the average
club level competitor.
So Andrew this is something that we've seen becoming
more and more prominent with World Time Attack cars
as well as motorsport obviously all around the world.
Can we start though by going into your background
a little bit,
how did you get involved in aerodynamics?
- I kind of fell into it.
I'm not actually trained as an aerodynamicist,
but as a mechanical engineer and then I was in motorsports
and then we started just finding lap timing,
these little things and I'd always loved fluids
and you know had a reasonable aptitude with computers.
So I just started experimenting and learning a lot
and then it became my profession slowly.
It turned from a hobby to working with other professionals
and then just became a full time business actually.
- So if you want to get into aerodynamics these days,
are there specific courses that people can take
at university level to train or do you have to sort of
self train if you wanna look at motorsport aerodynamics?
- Yeah no so there's a couple of respected aerodynamicists
around here that run courses.
We had one in Australia but we haven't had a venue
for it for a little while but there's a guy named
Scott Beeton who has an online one,
you can search him out, AeroDesign.
And then also Sammy Diasinis,
I think I pronounced that right.
He's a lecturer at I think University of New South Wales.
He's a designer at Porsche and they're both solid guys
and so those seminars are definitely out there.
There's plenty of books on the subject
although they're not as in depth as you know
the knowledge that's on the high end,
they're good introductory courses.
- So as essentially you got into aerodynamics,
and found speed in the cars obviously you've developed
that angle and I think this is something that maybe
is overlooked when it comes to getting speed and lap
times out of cars, there's a few areas we can go into,
obviously engine performance is generally an easy one,
there's also working on suspension
and mechanical grip from the car,
and then of course we've got aerodynamics,
obviously if we want the perfect result,
we need all of those working together.
But in terms of like factoring the importance
of aerodynamics from your view.
How does that sort of rate with the other two areas
in terms of what's most beneficial?
Like if you had to choose one of those three,
what's gonna give you the best improvement
from your car?
- Yeah that'll vary a lot car to car and team to team
because how good do they have each of those categories?
But they do enhance each other.
You know like the suspension vehicle dynamics
is a magnifying lens for the aero and you can change
how the aero works so much with the vehicle dynamics.
So understanding those interactions is really important.
At this level I mean as you saw in Time Attack,
if you look back in 2009, 2010,
nobody was under one minute 30 here.
And the basic specifications on the cars have not changed
a lot in terms of their, maybe some of them got lighter,
some have got heavier because they added aero where
they didn't use to.
Suzuki for example is more than 50 kilo heavier
than he used to be with aero but much faster.
But the whole field's moved into the low 20s from
the low 30s and I think aero is one of, if not the major
factor behind that.
The tires have been the same,
actually they started making more power too recently,
Billet engines have made a difference.
I mean that's not adding up to nine seconds or whatever.
- I think there's probably a general belief out there
in the market that if you want to develop real aerodynamics
packages for your cars, you're looking at spending
huge amounts of money,
and this may be off-putting I think.
Obviously aerodynamics when we think of it from
a layman's perspective, we're looking at the F1 teams
and the sort of budgets these guys have.
But it's possible to actually achieve a really effective
aerodynamic package, even at a club level,
would that be fair financially?
- Yeah I mean I don't think that's the case at all
and I think when we first started doing aero for Time Attack
it had that misconception and people were like
oh this has all gone too far.
And I completely disagree with that because compared
to what teams spend on engines I think aerodynmics
certainly can be inexpensive and I can give a few
concrete examples of that.
You know like the first Time Attack car that I ever did
was this NSX in the States that ended up becoming
the champion and taking this year's serial lap record.
And that car had an aero build budget of I think $350
of sheet alloy in a shed and we ended up going
3.5 seconds quicker like that.
There was no CFD it was just like what the team knew
and what I knew and our fabricator didn't sleep very much
and that's what we came up with.
So I think you first have to wipe it out of your head,
that there are shapes that have to be perfect,
like a wing is a really fine tuned shape,
and then there's some things that are like taking
a sledge hammer to the car and making a really big impact
just because people don't understand how it works,
and bringing that knowledge level up,
training the teams about how to use that aero
will make a big difference as well.
So I think another example would be like Nick Ashwin,
or Under Suzuki, or even Andy Forrest
where they've built their own aero.
So they've taken so much of the cost,
like the raw material cost is not that much.
And the design cost is not that much,
it's getting it built that's expensive.
And if you take on that attitude,
I think composites are new to this group.
Everyone's been fabricating for a long time.
Like Andy Forrest and their team, they got on YouTube,
and they got videos about how to make carbon fibre.
I got a long list of teams that learn how to make
their bodies on YouTube.
- So I think the point here is AMB Aero,
you offer a consultation service.
So it's not a full house,
or it doesn't have to be a full house service where
a team drop off a car, come back, write out a massive check,
and pick up the car finished with a full aero package
attached, you can work in a multiple different ways
including just providing some consultation,
telling the teams what they need to do,
and then allowing the teams to actually implement
those changes themselves and that can be quite
a cost effective option?
- Yeah I mean we have an entry level package
designed for those teams, that was the point
of what I wanted to do when we founded this business
with my partner was that we wanted grass level
people to be able to do aero,
and for it not to be this thing where they've built
a bunch of stuff that didn't do anything.
But to give them a proper aero build of some kind.
And we wanted to do that for every budget.
So we start at you know $2000,
we have packages for $2000.
There's a customer that had made a video
that's out there on the internet as well.
If you search about that, you can find a lot of stuff.
But then we've also got mid range stuff,
you know $7500, $15000.
If you wanna be in the Under Suzuki, MCA, Andy Forrest
level stuff, yes that's a significant design cost.
But still pales in comparison to production.
- I think straight away those numbers you're talking about,
starting at USD$2000 for some consultation
is probably a lot cheaper than most people are thinking.
I mean that probably is comparable to what a team
is spending on a set of tires and you're getting a real
world advantage from an aero package that then
they can go and implement themselves.
Let's talk a little bit about your actual design process
when you're faced with a fresh car,
one that you haven't working with before.
So what's the first step if you've got a team
that has a reasonable budget,
and they want you to develop something that's actually
gonna be really effective at maybe the pointy end
of the pro class here at World Time Attack.
- Yeah so I think the fundamental principle
of the design will all be catered around the team firstly.
Who's building it, who's driving it,
who's managing the program,
and it's quite custom to each of those factors.
Like we could do two of the same platform in a row
and the car will be very different because we think
very much about what's realistic for this team to build.
And if we go and design something that an F1 team
can manufacture, and it's some guy in a shed
with his mates, that's not gonna happen.
So you have to be I think really diligent
about catering it that way and that's why you see
such a big diversity of our packages,
like you'll see us go from nemo to scorch
and they're so different.
You basically really customise it to those factors.
And we can even tune the aero to be more or less
sensitive so it's got more peak downforce
but harder to drive because a pro might be able
to do that but an amateur cannot.
We have some many things we think about that way.
- In terms of dealing with developing that package
and giving the team a model to work with
or the shapes to work with and sizes et cetera,
you're starting by actually digitising the car, correct?
- Yeah, yeah so with you know,
the entry level package is more just what we call
best practice, is where we look at the car,
we help them figure out how to make sure it cools
and works better aerodynamically.
And then from the middle package on up,
which we call our Time Attack Pro Racer package,
obviously we have packages outside of Time Attack,
so those are where we start scanning the car.
So usually I'll fly in myself 'cause I wanna spend time
with the team and teach them about aero.
And at the same time we do that,
we 3D scan the car and then take that back to the office
where we process that and turn it into the 3D model
that we use for CFD or wind tunnel,
or whichever thing they're gonna do.
So talk to us a little bit about CFD because that's
something that's obviously become more prominent
as computing power has increased.
Obviously reduce the cost and development time
of aero so how does that process work?
- Yeah that's actually, I mean I might be going off topic
a little bit, but that's been so fascinating to watch
because the way computer power has changed
over the last 10 years is what's made that now accessible.
So that's really changed the landscape of CFD now,
because now these small companies can spring up
that are able to do this high end development,
that used to be totally impossible,
'cause you had to have a data centre sized super computer
to do, to even dream of it.
And do that's what's possible now and we've become
half an IT company in a way because we're just constantly
building machines and maintaining them.
But CFD's very powerful, it has its weaknesses,
but it has its strengths that the cost is there,
and if you can dream up a shape,
you can draw it and you can test it,
you don't have the physically build this thing
in real life like you do in a wind tunnel.
- So massively reducing the development cost
because when you finally go through the process
of making a physical part,
you can be pretty confident from the CFD results
that that part is going to give
you the results you're expecting?
- Oh yeah and the quantity of tests you can do.
So I mean there are some things that are faster to do
in a wind tunnel like adjusting a wing angle for example,
you know CFD that's another run.
But we could test so many parts in a week
in CFD that you could never dream to do in a wind tunnel.
Just stuff that's out, you think of something out there
and you learn from it and that's still impossible in a wind
tunnel without a team of 50 model builders.
If you tried to match I think how many crazy things
we could test in a day,
you'd need a hell of a model team in a wind tunnel.
- In terms of validating the results that you're getting
from CFD, when you do have the opportunity
to work with a team that has a budget for wind tunnel
testing, I mean typically how well do the results
from the CFD analysis line up with real world
downforce results?
- I would say that our CFD is lining up probably
with one exception that we're still getting
to the bottom of,
but with the exception of that one car,
historically we've been closer than the wind tunnels
were to reality.
So far, most wind tunnels.
Now there are very good wind tunnels out there,
you know $20000 a day wind tunnels,
but there's a lot of wind tunnels that are so so.
But those things are actually not important and we
don't worry so much about that unless
we find a hole in what that wind tunnel sees,
we look at relative accuracy, not absolute accuracy.
Absolute figures, I believe should only be derived
from a measurement on the real car.
We look at a CFD gain value and we've learned to trust
that and there are some things we've learned about CFD
we know yeah that's kind of the boundary of CFD.
But the wind tunnel's the same way,
there are certain things the wind tunnel
will tell you wrong.
And there are times when they'll contradict each other even.
And having that depth of knowledge,
that experience, you can just stay away from that kind
of a design and that doesn't mean you're giving up
something because you just spend your time
developing other things you know are accurate.
- I wanna delve a little bit further into the wind tunnel,
but first of all if you aren't using the wind tunnel
to validate your designs,
how do you validate those in the real world,
how are you getting the results off the car
at a racetrack?
- Yeah well we do do wind tunnel validation
if the customer budget allows.
But most time attack teams,
we've only had a few exceptions to this,
did not have the budget for CFD to wind tunnel
to on track and we were the first to do all three together
but that's quite rare.
So we do rely a little bit on the data we've gathered
from one car that's managed to do that
sort of correlation project
and we have to sort of apply it everywhere
because the budget just doesn't allow for that.
But we do push really hard for all of our teams
to instrument the car at least in some way
so that we can validate the data.
- So we're talking here load cells in the suspension system?
Yeah load cells, shock pods, ride height sensors.
And there's ways, like I mean there are really primitive
ways that you can do this if you've really not
got a budget,
I mean when I started out racing my own car
and Bonneville stuff,
you know we'd go out on the dry lake
and we stuck zip ties on the shock shaft
and did a coast down test.
It's rough, it's rough data, but data's data.
We were able to make a car that was unstable
stable that way.
So that's a win and so I don't think,
I would never tell a team to not test because you don't
have the ultimate way to test,
you've gotta validate any way you can afford to do.
But it's not really that expensive in the picture
of building a car to put four shock pods on there.
But the really important thing is the process.
I cannot stress enough how difficult it is to do
that process properly.
Even if you have the sensors,
we get customers come back with data,
and they're like oh the downforce isn't there
and then this happened really recently with a customer,
and we've just forced them to put it on the scales
and then load up the car incrementally
and then see what the actual curve laid out as,
and it was drastically different,
it actually lined right up to CFD
where for a year they'd thought
they were missing downforce.
- So essentially if you don't know how to analyse the data
that you've got then you're just guessing.
- So that's a big part of what we try to do is to get teams
to measure as good as we can.
It's ideal when you can sit with them through the testing
but that's not always realistic so we try to train 'em
and we give documentation to them about
how to do that the best they can.
- OK let's just move back to the wind tunnel testing
because obviously that's another area where we've
grown up seeing that being sort of related to F1,
that sort of level.
But you were talking to me earlier about model,
part scale testing that can be affordable for the teams
who are at the pointy end of a semi professional
motorsport such as World Time Attack.
Can you talk us through how that works?
- Yeah so there's a technology that we kind of,
I hope we pioneered it,
I dunno who else is doing this.
I mean Suzuka, my partner slash mentor,
had this idea that most of motorsport was using
too big of a scale for what they were doing.
And I think that this was all based on what markets
they're selling to.
Like if you've got, OK 50% scale wind tunnel,
60% let's say, that's a standard kind of scale wind tunnel.
- Just to step back there,
so for those who aren't aware,
what we're talking here is a scale of the size of the car,
so a 1:1 obviously we're taking the actual car,
running it in the wind tunnel,
but that gets very costly,
the wind tunnel needs to move a huge amount of air,
hence the wind tunnel becomes more expensive,
so it becomes cheaper if you're using a scaled down
tunnel with a scale model of the car,
so that's what we're talking about.
- Yeah exactly and it's the cost of building the models
and the parts to test.
If you went to a full scale wind tunnel 1:1
You have to physically build in full scale,
every single part you're going to test.
How many parts can you test in a day?
That's really hard, and you have to plan out ahead
to build all the parts you're gonna test.
You can't be guided by testing,
you can't get feedback from the testing.
And so that becomes very very difficult to do.
Now if you move to smaller scales,
you start getting new technologies coming in,
such as 3D printing.
You can 3D print parts,
they take less time to construct for various means.
They don't have to be as rigid,
because they're taking less load in the smaller scale.
So the cost of wind tunnel goes up at the cube of the scale.
So it's really a high, you know it gets really nasty
at the high end.
- And when it comes to that small scale testing though,
the accuracy of the model becomes more and more
important.
You were talking to me earlier,
one of the issues, one of the important issues
with aero is the ride height,
and of course when you scale down the model,
those ride height changes become more and more precise,
so can you tell us how that works?
- Yeah I mean every single dimension on the thing
becomes more and more high precision as you get
to the smaller scale.
So that part's difficult.
And that is part of what you accept as your plus/minus,
allowable plus/minus and how true you think
this model shape is to the real car.
Well you know we integrate 3D scanning back in,
so we have like a smaller scale,
really high accuracy scanner,
we scan the models back in so we get a better feel
for how far off these things are.
But you still struggle with those
kind of things at small scale.
You know what the problems are,
what the shortcomings and the difficulties are,
and then you use it for what it's good at,
and when it's not good anymore you move
onto the next thing,
and that is a much more efficient process
if you do it that way.
- Now you mentioned that some of the large 1:1
scale professional tunnels,
you're sort of talking $20000 a day or more,
if we're looking at some of these cheaper
scale wind tunnels,
can you give some indication of what a day testing
on that wind tunnel might look like?
- You mean like cost wise.
Oh there's a bunch, like I dunno in Australia,
but I know like in North Carolina
they have some of these ones like the NASCAR guys use,
and they're really inexpensive.
You know you could go there for like two or four grand.
But there's no moving belt in them.
So there are some things you can trust
and some things you can't.
I mean you don't have a moving belt.
If your team is looking at trying to make this
figure out the cooling system or the rear wing
you know you might make major imporvements
in a tunnel like that so you never rule it out
and say oh that testing's bad testing, forget it.
But you gotta know what it's good at.
- OK I wanna move on and talk a little bit about
the aero balance of the car.
So I mean again from a layman's perspective
it seems like it'd be a relatively easy task
to go and put a big wing on the back of a car
and make a tonne of downforce at the rear of the car,
but of course that's going to affect the balance of the car
because we've only looked at the rear.
So obviously in order to make the car work correctly,
it's important to balance the downforce from the front
to the rear.
So how do you go about doing that and obviously,
CFD you've got some ideas there,
what can you do to allow that to be adjusted
in the real world to suit the car and the driver?
-Yeah I mean there's so many strategies
for how you do this and we cater it a lot to the car
and the team but the basic concept in Time Attack
is we give them an adjustable rear ring,
and then we have a CFD value that we found
that most drivers are happy with and we sort of float
around that percentage.
And it's based around the weight distribution of the car.
So there's a percentage variance from that
which is a driver preference or a team preference.
There's a lot of complex factors going on
like how the car is actually
changing its attitude dynamically.
Like what it's doing in yaw, roll, pitch,
and there's so much complexity to that and we're learning,
we learn every day about this, today especially.
We're still learning right.
And that's the point why we come out here
is to try to find out those kind of things.
But we get that percentage smaller and smaller
every time.
And the tighter we're able to make that adjustment window
as we learn more,
then the more we size the wing appropriately
and then they end up being more efficient.
That's one of the ways we can make the wing drag less
because we don't have to make it able to have an extra
20% rear we go no no this is fine
tuned right for this number.
- OK the other thing we see with the high downforce cars
is it adds a complexity to the suspension design
because you've got a car that has obviously zero
downforce when it's sitting stationary,
as the speed increases the downforce also increases
which tends to compress the suspension
so particularly when we're getting cars here
that are getting sort of 280 kilometres an hour
at the end of the front straight going into turn one,
this compresses the suspension down.
So how are the teams best to deal with this?
Do we run higher spring rates and compromise
the suspension system, run it on the bump rubbers,
or third spring and damper set ups,
what's your preference?
- Well to me I mean obviously a third spring
is a much better situation because you have
a heave control system.
And that's really what you'd want,
I mean short of an active suspension
or something like what Andy Forrest has got
where he's got an actual,
you know he's affecting the length
of the push rod dynamically on the third.
- Yeah no we looked at Andy's car.
He's got a little air canister,
I don't believe that he's actually using it yet,
but ultimately it will be able to change the ride height
at speed essentially.
But yeah in most instances the third spring or heave spring
is the best arrangement.
If we don't have that, what's your second best?
- The second best is gonna be,
you're gonna have to use some combination of bump stop
and spring right.
Because what you've got is if you make the car rigid
enough to endure turn one without using
any bump stop whatsoever.
Then with the kind of air load's we've got now,
you're gonna wind up with this spring rate
that's way more than the tire spring rate.
Which means that all the energy being fed into this car
around the racetrack is going into the tire.
And you see the teams now are working on trying,
some of the teams that have more downforce,
are working on trying to ride
this limit of tire versus failure.
And either one of those things will contribute to it.
Because they both feed energy into this tire,
and cause it to overheat or to damage the side wall
and come apart.
So with the third spring set up like Andy's,
that's just letting you control that so you're able
to make a car that's softer for a greater percentage
of the lap and the time when
the car is really rigid is less.
And that's the basic concept.
With bump stops, you can do that, but not as well,
because inherently, let's say you tuned it,
you said OK I wanna use the bump stops
only on the straight away,
then you can, you have to have a ramp rate,
like you can't make it just instantly
go to one or the other,
the car's difficult to handle.
So you now like high end of motorsport,
you'll sit in a room full of engineers,
and this is all you do all day,
is looking at an air map,
and looking at the suspension and going,
how do we get more aero out of this car
and not blow up tires,
and not make it difficult to handle,
and that's how prioritised aero is at the top level.
And this is becoming that way.
So there's no solid answer about this,
it's gonna be, you've gotta get somebody good
that knows when they're doing with that stuff,
and they're gonna be constantly tweaking that
and changing it track to track.
- So essentially, long story short there,
we're always going to be focusing on a compromise
if you wanna get that aero and make the suspension
and the aero work together?
- Yeah, yeah and then you've gotta have good aero too,
like if you have no idea what the aero's doing,
what are you doing with the suspension?
So it's important to make that work as a package.
That's what the top end of the field will be doing
moving forward and you see that,
you see that happening now.
- Another common myth, or I dunno if it's a myth,
I wanted to get some answers on this,
another common thing I hear is that aero doesn't really
produce any downforce until the car's perhaps up around
100 mile an hour, sort of 160 kilometres an hour.
Is there truth in that?
Obviously the aero downforce does increase with speed,
but what sort of speeds can we start getting really
useful increases in downforce?
- That depends on how much downforce you've got.
Right if you've got a car that's got mega, mega downforce
in the slowest corner on this track,
so like our highest downforce track out here today
in the slowest corner here,
has twice as much downforce as a GT car does
in a high speed corner, you're gonna feel that right.
And if you look at the trend,
people have that misconception,
and the evidence I would give to disagree with that
is if you look at the top end autocross hill climb,
all those cars, massive, massive aero.
And it's really important for them.
So yes the speed's lower but then they've put
low speed aero on it and it's just as important
as anywhere else.
And if you look at even really tight courses,
you look at like Formula SAE.
Those are the tightest courses there are.
Super low power, super tight courses, mega aero.
And it's a percentage of the lap time.
For them if they gain two, three tenths on aero,
it's the same things as gaining a second and a half here,
'cause the course is only 30 seconds long.
- So essentially the speed,
you're just designing the package to suit the speeds
that the car's gonna be seeing.
- Yeah just changes your drag curve,
and we look at that in simulation.
As we go through our test runs and we look at it
in our database and then we see how much drag
we're picking up and we run simulations
on the change in downforce and the change in drag
and we make sure that we're always on the right
side of that curve.
And sometimes we try to be a little low drag
more than that computer predicts,
we think that's the right way.
But you have to calculate these kind of things.
- Look Andrew thank you for the insight there.
I think hopefully that's dispelled some of those myths
that proper aero is only in the league of those
with million dollar budgets.
And if people do wanna find out more about you,
maybe work with you, how can they get in touch?
- Yeah you can just go onto our website,
it's amb-aero.com
- Alright thanks for the chat Andrew.
- Yeah no problem.
- If you liked that video, make sure you give it a thumbs up
and if you're not already a subscriber,
make sure you're subscribed.
We release a new video every week.
And if you like free stuff, we've got a great deal for you.
Click the link in the description to claim your free spot
to our next live lesson.
You'll learn about performance engine building
and EFI tuning and you'll also have the chance to ask
questions which I'll be answering live.
Remember it's 100% free,
so follow the link to claim your spot.
For more infomation >> JUDGE KAREN...THIS CHILD RIGHT HERE,"OH MY GOD".."MAN LEAVE DEM DRUGS ALONE"...THAT FAN THO!!! - Duration: 3:42. 
For more infomation >> Government Agency BUSTED Lying To Undermine Trump! Here's The Truth They're Hiding - Duration: 5:19. 
For more infomation >> 'I'm So Grateful To Be Here' - Duration: 0:38. 
For more infomation >> The Voice 2018 Blind Audition - D.R. King: "Believer" - Duration: 4:28. 
For more infomation >> Dolly Parton Dedicates Her Imagination Library's 100 Millionth Book to the Library of Congress - Duration: 40:25. 
Không có nhận xét nào:
Đăng nhận xét