Red Bull RB7 Renault

[PlatinumZealot]

Yes, but you missed my two points.

3. My point one more time - even if the rake is only 1 degree, downforce from the floor rake alone is worth tens of kilograms and every team on the grid knows that, and considers it significant in car's design phase. None of them even try to run CFD and wind tunnel without some rake. RB7 just has a little more then the others.

So why did you use that graph of a single airfoil to characterize the whole car then? I don't think it can account for all the ground effects and flow conditioners and obstructions on the car - so I don't think you can extrapolate that linear relationship to a whole of a car just like that.

[TURU]

The formula for a line is:
y= mx + b
You left out that b.
b is around 0,35

let’s take two points to prove its linear.
(5/0,7) and (10/1,05)

Let’s calculate m with first point:
0,7=m*5+0,35
m=0,07

Let’s check now if second point is located on that line.
y= 0,07*10+0,35=1,05
You see it fits very well.
Therefore it’s a line.

Sorry, but your proof is no proof. Firstly, how can you derive b to 2 decimal place just by looking at it? Then, you can't take 2 random points out of the graph and choose their coordinates so that they fit your assumption. And even if they did, it still wouldn't prove that all other points lay on this line too.

[SiLo]

Back on topic...

If they have to move around all the KERS pieces, do we think they will have an even longer time trying to get it right over just fixing it where it is?

I'll just leave this here.

[i70q7m7ghw]

Wouldn't some teams then, "design" some rake into their cars? Eg. Take a sideways drawing of a McLaren or a whatever F1 car. Instead of having the floor level with the edge of the paper at that perspective, imagine drawing a line which is lower towards the front of the car and higher towards the back (ie the "stock" setup would already have rake)

That's what Newey did. I'm surprised others dismissed it.

All F1 cars are designed with a small amount of rake, it doesn't take much 'negative rake' to create lift from a flat underbody, think back to Webber's little flight last year. For that reason a modern F1 car will not have a underbody perfectly parallel to the ground. It does appear as thought Red Bull have taken the idea to an extreme and run alot more rake than others.

I'm not sure how much this whole engine/gearbox/etc. designed at an angle - or not so - is relevant? Ferrari ran alot of their packaging at an angle last year to fit the double diffuser in, it didn't seem to stop them from challenging for the championship?

[Mr Alcatraz]

Back on topic...

If they have to move around all the KERS pieces, do we think they will have an even longer time trying to get it right over just fixing it where it is?

I'll just leave this here.

That is a question that has already been asked a few pages back. IMO I think Red Bull is determined to make it work where it is, due to how well the car is balanced. That is the impression I get, and it is supported (somewhat) by Webbers comments I've placed in quotations below.
FWIW I think they could get it to work in a more conventional configuration, but that is not how they designed the car

"It means that we have to work harder, we have to work as cleverly as we can and as fast as we can. Fortunately, we now have a bit of a break, so it will be a clear focus for us in terms of durability and consistency and also understanding the system.

"It's not only a performance thing but it's also incredibly disrupting in the garage. My car was rebuilt about four times this weekend," he added.

"It makes it so much harder for them in terms of making mistakes, when you open a car up and put it back together all the time, it's not easy."

http://www.crash.net/f1/news/168625/1/r ... mpaign=rss

[i70q7m7ghw]

That's no a fault of KERS, that is a design fault by Red Bull.

RB constantly bashing KERS is getting a bit boring, they need to learn how to build a proper system, rather than whining about it and trying to get it banned.

[mep]

Sorry, but your proof is no proof. Firstly, how can you derive b to 2 decimal place just by looking at it? Then, you can't take 2 random points out of the graph and choose their coordinates so that they fit your assumption. And even if they did, it still wouldn't prove that all other points lay on this line too.

You are very nitpicking.
This is a linear approximation to possibly somemeasured values. Even if some of the points don’t fit 100% to the line it still remains true. Don’t forget that there are measurement tolerances, too. Even if one point would lie completely outside of the line we could still draw this conclusion because this single point could be declared as measurement failure. That’s how engineering works. It’s therefore different to the way a strict mathematician would think. Some failure tolerance is allowed here. That’s why the linear approximation is correct.

Btw: You can read out figures up to second decimal when the annotation of axis has already that accuracy.

However you can’t seriously question the linear behavior of the profile in the range between -5° to 15°. It’s really ridiculous to argue against that.


Just as comparison wesley123
1.failed to read the diagram properly
2. used the wrong math
3 failed to read out the points correct.
And by this decided the profile has no linear behavior. If you prefer to believe him, have fun.

[wesley123]

Okay as you say for the formula;
y=mx+b that in first isnt even correct as x is the one you fill in, it is an variable, a not defined part of the formula. It doesnt matter at all for the outcome but it gets kinda messy when filling in. so we rather take y=xm+b where m is the raise in Coef per AoA step. b is the starting point with AoA of 0.

Lets fill these in at AoA of 5 which is x here gives an y of 0.7.
at an x of 0 y is aproximately 0.35 that means that with an raise of 5 AoA the lift coef riased by 0.35. so per AoA raise that means that Lift Coef raises by 0.07

Now lets fill these all in

y=x*0.07+0.35

Now let's try a few points; We see at an AoA of -5 the lift coef is 0. so lets fill -5 in for x.
y=-5*0.07+0.35
that makes y=0
and let fill in x=10;
y=10*0.07+0.35
y=1.05

Now you see, it is indeed linear. but what is the point of an linear graph, that is right it is linear, at x=18 we see something weird happen, the graph bends down slowly. Therefore the graph is not linear. Sure the graph before seems.

Oh and btw check this; added a straight green line, you see the blue line go out of the green line and then go back.


Not quite usual for a linear line to have a bend in its line is it? and even if i put the line wrong, the blue part that shows would have been constant and not return to the green line.

So the line is not linear, it is close but it is not linear

[volarchico]

As far as real-life engineering goes, that line is as close to linear as you will find. The real world is never perfectly linear due to measurement errors or (gasp) non-linear effects. But it sure is easier for engineers to estimate using the linearity assumption...especially when they realize that having a 0.1% error will not grossly affect the outcome.

[wesley123]

that is true, but on that graph i think it is a bit more then that 1%, i think it is closer to 5%, in F1 1% is already way too much off, what do you think if this graph was the start of the whole research, over the whole research you get further off and off just because of this mistake.

Apart from that, that was a graph of a flat plane, when taking a complete car this will be even further away from linear, therefore, although it is really close to being linear, i am in the opinion that you are not allowed to call this line linear due to the simplification of a 640 kg war machine

[ringo]

I am of the opinion we are better of reading this graph,

than that one.
That's how related to the topic that airfoil graph is.

Unless for a mental exercise, it's misleading to infer anything from that airfoil graph.

[flyboy2160]

I am of the opinion we are better of reading this graph,

than that one.
That's how related to the topic that airfoil graph is.

Unless for a mental exercise, it's misleading to infer anything from that airfoil graph.

I just discovered this great website a few weeks ago and this is my first post. I am using it to say that regardless of technical ability, I think ringo has the best soh on the site. Keep up the CFD and wisecracks!

edit p.s. And in this case, he is not only funny, but also correct! That airfoil aero stuff has nothing to do with the messy boundary layer vortexy flow under the car....

[mep]

[...]

Now you see, it is indeed linear. but what is the point of an linear graph, that is right it is linear, at x=18 we see something weird happen, the graph bends down slowly. Therefore the graph is not linear. Sure the graph before seems.

At 18° the profile stalled. It should be clear that this will happen at some point. However initially marekk was talking about the rake of cars so we are talking about angles between 0°-5°. Again the behavior can be described very well as linear in this area.

Apart from that, that was a graph of a flat plane, …

The graph is not for a flat plane it is for a Clark Y airfoil. The Clark Y is a very often used airfoil so you should already have heard about. Even if you don't know the profile you could have seen on the graph that it is not a flat plate because it already creates lift with negative angles of attack.
Here a picture: http://www.ae.illinois.edu/m-selig/ads/ ... clarky.gif

[N12ck]

[...] How do they actually determine the 'rake' of a car, as i thought the reference plane was the parallel between the wheels, so surely that would be breaking the rules for height of the car unless the reference plane is the floor,

[marekk]

Now as we all agree that even small amount of floor rake produces significant (for F1 i define significant as > 0.1%) amount of downforce and all teams are considering this during car design and CFD/wind tunnel runs let's go back to topic and try to understand why higher then average rake works for RB7.

Even small change in rake angle cause significant changes of front wing ride height (1cm for every 1/4 degree, more or less).
I'm not gonna to place new graphs of wings Cl and Cd in ground effect to discuss this, so just let's suppose McLaren's estimations of 1cm lower FW ride height are worth > 0,5s of lap time are true.

That's easy part - every team could just dial more rake into their setup and immediately gain FW efficiency.

Tricky part is at the rear. More rake means more rear ride height and more air volume/mass under the car one have to accelerate (flow speed under the car is where floor downforce comes from), so bigger pressure gradients needed to move this bigger mass of air at the same speed as on less raked cars.
Geometry of floor/diffuser/beam wing combo is restricted by the rules and doesn't vary much between teams.
The key differentiator will be then how much energy the flow to the gurney and beam wing has, which can then be converted to low pressure behind beam wing and to create strong vortexes behind gurney (2 of them, counter-rotating along rear gurney wall, lower one helping to extract more air from diffuser).

McLaren and maybe TR have good potential to go this direction with their designs, but in RB7 this concept has already 3-4 years of development history.

The biggest problem i see with this is concept is there is no real room to improve it further. Even KERS integration seems to be out of reach at the moment, and those big cooling outlets are already a compromise. Newey not even tried to look closely at Renault's solution (biggest potential IMO) - there is simply no space in RB7 left to accomodate this.

@N12ck: reference plane is the floor, you can run as much rake as you want.
@ringo: not as funny as CFD, but LOL anyway. Thanks.