The 60%, 50m/second rule...

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
During design and testing, Formula 1 teams are only allowed to build and test a 60% scale model of their proposed car design and run it in a wind tunnel at a maximum speed of 50 metres per second (about 110 mph).

How do they compensate for the difference between these testing models and the larger and faster figures in the real world of an F1 car?

Does CFD fill the gaps? Is it simple maths and multiples?

Jonathan Neale of McLaren says that they are literally addicted to the wind tunnel. So will the wind tunnel approach prevail in F1 design forever?
 

cider_and_toast

Exulted Lord High Moderator of the Apex
Staff Member
Premium Contributor
I wasn't aware that those restrictions were in force. I wonder how they police that?

As far as wind tunnel use goes. As long as Aerodynamics remain the key to a fast car then there will always be a need for a wind tunnel. The only way I'd ever imagine it would go out of use is with mandatory aero packages and god forbid that ever happens in F1.

I guess CFD must fill in the gaps and then some clever maths does the rest.
 

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
I also wonder how they make allowances/test for cross winds etc. instead of just testing the car head on...in fact I wonder if they can simulate the sort of turbulent air that comes off the back of an F1 car as well?

So many unanswered questions! Bah!
 

snowy

Champion Elect
It may save money in the modelling dept but there's a host of nearly brand new full size wind tunnels being wasted! They sit about 90% of the time doing nothing and 10% testing half-assed cars.
 

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
I wasn't aware that those restrictions were in force. I wonder how they police that?

As far as wind tunnel use goes. As long as Aerodynamics remain the key to a fast car then there will always be a need for a wind tunnel. The only way I'd ever imagine it would go out of use is with mandatory aero packages and god forbid that ever happens in F1.

I guess CFD must fill in the gaps and then some clever maths does the rest.

Welcome to your worst nightmare (according to F1 Racing [Pat Symonds])...

http://www.pistonheads.com/news/default.asp?storyId=23084
 

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
The article is based on an interview with Pat Symonds. F1 Racing magazine is available from all good retail outlets.

Jonathan Neale’s interview where he states "CFD is important to us, but we’re wind-tunnel junkies…" can be found here:

http://www.formula1.com/news/interviews/2011/2/11766.html

Also, here are the official FIA rules on wind tunnel testing I discussed in my original post.

From the Federation Internationale de l' Automobile:

22) TRACK AND WIND TUNNEL TESTING
22.1 a) Track testing shall be considered any track running time undertaken by a competitor entered in the
Championship with the exception of :
i) promotional or demonstration events carried out using tyres provided specifically for this purpose by the appointed supplier ;
ii) one three day young driver training test, carried out on a site approved by the FIA for Formula 1 cars and between the end of the last Event of the Championship and 31 December of the same year, any such driver having not competed in more than two F1 World Championship Events ;
iii) four one day aerodynamic tests carried out on FIA approved straight line or constant radius sites between 1 January 2010 and the end of the last Event of the Championship. Any of these days may be substituted for four hours of wind-on full scale wind tunnel testing to be carried out in a single twenty four hour period.
b) No competitor may carry out more than 15,000km of track testing during a calendar year.
c) No track testing may take place between the start of the week preceding the first Event of the Championship and 31 December of the same year with the following exception.
If a team declares that one of its current race drivers is to be substituted by a driver who has not participated in an F1 race in the two previous calendar years, one day of track testing will be permitted between the start of the week preceding second Event and the last Event of the Championship. The following must be observed :
- Any such day may only be carried out by the new driver and may not take place on a circuit hosting a race in the current Championship year.
- Any such day may only take place within a period 14 days prior to the substitution and 14 days after the substitution has taken place.
- If a team, having declared the driver's substitution and performed the test, does not then enter an Event with the new driver, the team will be penalised by a reduction of one day from the pre-season track testing days available in the following year.
d) During all track testing cars must be fitted with the FIA ECU required by Article 8.2 of the 2009 FIA Formula One Technical Regulations.
e) No track testing is permitted at sites which are not currently approved for use by Formula 1 cars. In order to ensure that venue licence conditions are respected at all times during track testing, competitors are required to inform the FIA of their test schedule in order that an observer may be appointed if deemed necessary.
f) During all Formula One track testing :
- red flag and chequered flag procedures must be respected ;
- no other type of vehicle is permitted on the track ;
- every reasonable effort should be made to ensure that the recommendations concerning emergency services detailed in Article 16 of Appendix H to the Code are followed.
g) If, after an incident during track testing, the Medical Warning Light signals that threshold forces have been exceeded the driver must present himself for examination in the circuit medical centre without delay.
h) With the exception of the full scale testing permitted in 22.1(a) above, no wind tunnel testing may be carried out using a scale model which is greater than 60 percent of full size.
i) No wind tunnel testing may be carried out at a speed exceeding 50 metres/second.

They can also be found here: http://www.formula1.com/inside_f1/rules_and_regulations/sporting_regulations/8713/fia.html
 

Grizzly

Bear
Contributor
I guess CFD must fill in the gaps and then some clever maths does the rest.

Well, in fact CFD is the maths...

I also wonder how they make allowances/test for cross winds etc. instead of just testing the car head on...in fact I wonder if they can simulate the sort of turbulent air that comes off the back of an F1 car as well?

So many unanswered questions! Bah!

You can simulate anything. Literally anything. The limits are your computing power, how much data can you simulate?

CFD theory is exact. However the data you give it and what you can do with it, is an estimate. Everything can be defined and fluid motion was defined many years ago, i don't remember the mathematicians name, I don't think it was Newton, but in/near Newtons time... hopefully someone can remind me, but anyhow...

For the last however many years they have remained fancy theories, it is only in recent times that computers have been able to run some data though these equations, but they are nowhere near powerful enough to simulate all the variables. That is why real world / wind tunnel testing will be with us for many years to come.

Yes you can simulate the air off the back of the car, yes you can simulate cross winds, but right now we have to select and reduce the variables/data inputs to get some useful results, this Millennia...
 

Grizzly

Bear
Contributor
This fun page from CFD Online shows simply the maths for calculating compressible fluids (gases, air) at a constant velocity. I.e. simplified and velocity changes completely removed...

Isentropic flow

Constant entropy flow is called Isentropic flow.
From a consideration of the second law of thermodynamics, a reversible flow maintains a constant value of entropy.
In the theory of streamtubes, isentropic flow is the basis for compressible flow.
The well known incompressible flow does not apply if
is violated. Therin

is called the Mach-number. It is the relation between the flow-velocity u (fluid or object) and the speed of sound a in the surrounding medium.

In a caloric ideal gas the speed of sound depends on the absolute temperature T (K), the adiabatic exponent
and the specific gas constant R (J/(kg K))
.
In this case all thermodynamic properties (temperature, pressure, density and speed of sound) can be expressed with explicit formulas which are functions of the Mach-number!

In the following all thermodynamic properties are related to their properties at rest (u=0). From the energy equation for a frictionless, adiabatic flow

you will get that the enthalpy at rest ht is allways the same regardless of an isentropic or non-isentropic state change. As a matter of fact -
- the temperature is it as well. In contrary the pressure depends on how the gas is brougt to rest. The pressure at rest is only obtained if the state change is isentropic. If the entropy changes the pressure at rest changes as well, e.g. when passing a shock.

Pressure





Temperature and speed of sound




Density



Critical values

Values which can be found at M=1 (speed of sound) are sometimes called critical values. They are marked with the superscript *.
They differ from the properties at rest only by a constant value and are therefore used as reference values as well.
In the special case for gases which contain two atoms (
) one gets:




Critical cross-section

The cross-section
in a nozzle where M equals 1 is often used as a reverence, even if M=1 is never reached.




Summary


As one can see all thermodynamic properties decrease with increasing Mach-number.

Whether you understand, have any interest at all or what, it is plain to see the amount of time and computer power you can donate to this is endless, and what can be simulated in a few minutes in a wind tunnel, can take infinite amounts of time via CFD. Literally billions of billions of calculations, and you may still be limiting the data/variables that you feed the computer in order to speed up getting some usable results....
 

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
Grizzly - Wow!

The mind boggles at all of that!

So, the upshot is that wind tunnels are currently more manageable I guess, or produce more useful/useable results, faster?
 

Grizzly

Bear
Contributor
I think its fair to say that in the last 10 years CFD has taken leaps and bounds forward with massive processing power and multiple processors now much more developed. The fact Virgin put a car on the grid last year is huge testament to that!

The upshot? Well i think they should be used to compliment each other. After all wind tunnel testing is a simulation just as CFD is too and the 60% 50m/s rule is designed to keep it that way too. We simply have much more experience with the fan and smoke...

In years to come the FIA will begin restricting processing power to prevent the 'he who has the biggest computer wins'. It might be 20 years from now, but it will come, as we learn and develop more, it will come...
 

Porceliamone

This cost me a tenner, but so L'Oreal.
Contributor
I think its fair to say that in the last 10 years CFD has taken leaps and bounds forward with massive processing power and multiple processors now much more developed. The fact Virgin put a car on the grid last year is huge testament to that!

The upshot? Well i think they should be used to compliment each other. After all wind tunnel testing is a simulation just as CFD is too and the 60% 50m/s rule is designed to keep it that way too. We simply have much more experience with the fan and smoke...

In years to come the FIA will begin restricting processing power to prevent the 'he who has the biggest computer wins'. It might be 20 years from now, but it will come, as we learn and develop more, it will come...

It's strange that the pursuit of excellence and innovation that embodies F1, the governing body constantly feel the need to reign in or dumb it down. I have difficulty getting my head around and accepting that sometimes...
 

sportsman

Sidecar racers have the biggest cojones
Contributor
Getting back to the original article there are a great many assumptions made that may or may not be true for 2020
Will F1 still be open cockpit. Maybe or maybe not.Will it still be open wheel as we know it today.Who knows.
Will groundforce effects generated solely by the monocoque be the only source of downforce and wings banned.
No one has any idea just what the regulations will be.
Even for 2013 there are rumours that limited ground force designs will be allowed.If that is true and the history of downforce generating chassis show that the "coke bottle" shape may not be the most effecient.
Its quite possible that someone will very carefully study future chassis regulations as and when they are published and come up with a revolutionary shape quite different to todays shapes.
 

Pyrope

Podium Finisher
Supporter
Interesting interview with Nick Wirth in the most recent Motor Sport magazine that goes into some of the complex issues surrounding CFD. A lot of the positions still seem to revolve around whether you believe that CFD can model turbulent (i.e. chaotic) airflow. Nick he say yes, Pat he say no.

Apparently the OWG (which Bent Pat was a big part of) used 40% scale models so that they could have two in the tunnel at once and examine how the flow from one car affected a following car. That wasn't exactly a resounding success, now was it? It has been a long time since I was professionally involved with CFD, FEA and other fun matlab applications, but I'm prepared to believe that CFD and wind tunnels are at least comparable these days. However, although Nick Wirth may well be a fantastic aerodynamacist, he's clearly an atrocious structural engineer. Until Virgin can get round that issue they are going nowhere.
 

Grizzly

Bear
Contributor
Interesting interview with Nick Wirth in the most recent Motor Sport magazine that goes into some of the complex issues surrounding CFD. A lot of the positions still seem to revolve around whether you believe that CFD can model turbulent (i.e. chaotic) airflow. Nick he say yes, Pat he say no.

Well, they are both correct and both incorrect. Yes, CFD can model ANYTHING, PERFECTLY!

No, the computers we have today can't model much more than, relatively speaking, basic, rudimentary, simplified simpleness.

Fluid dynamics is physics/maths, physics can be defined into basic laws, anything that can be defined can be calculated. IF you have the power to calculate it. We don't and we won't for years to come.

Apparently the OWG (which Bent Pat was a big part of) used 40% scale models so that they could have two in the tunnel at once and examine how the flow from one car affected a following car. That wasn't exactly a resounding success, now was it?

To be fair to the OWG, they have not done anything other than place models in a wind tunnel and mull things over. The ONLY advice they have given is to warn, that every proposed change that has so far been implemented or put forward in the name of increasing overtaking, probably won't have any, or very little effect.
 

Pyrope

Podium Finisher
Supporter
Fluid dynamics is physics/maths, physics can be defined into basic laws, anything that can be defined can be calculated. IF you have the power to calculate it. We don't and we won't for years to come.

Hmm, you are also assuming that we actually do understand turbulent flow. From what I have seen, we don't. Therefore any definitions that we have presently are only approximations.

The ONLY advice they have given is to warn, that every proposed change that has so far been implemented or put forward in the name of increasing overtaking, probably won't have any, or very little effect.

The narrow rear wing and snowplough front wing that the cars presently sport are direct results of the OWG's work.
 

snowy

Champion Elect
The simple solution to Virgin's CFD limitations would be to start a SETI type operation and get millions of people and their computers actively involved in computing their 2012 car. It would be a stroke of genius and very compatible with Richard B's business philosophy.
 

FB

Not my cup of cake
Valued Member
I won't pretend to be an expert in the area but I am constantly surprised that some very clever engineers think that CFD can ever replace real time testing in either a wind tunnel or on the track. I work in the aerosol industry and they have tried to model nozzle design to improve efficiency, noise etc. All the time the gas and liquid is contained with a channel it all moves very predictably but as soon as it exits into the atmosphere it pretty much does what it likes as so many other factors come into play.

I bet McLaren thought their 2009 car was a real world beater when they first modeled it on a computer and then stuck in a wind tunnel. As soon as they put it on the track it turned out to be a pup. As Pyrope points out the effects of turbulent flow are very difficult to model, you add into that wind speed & direction and humidity (both of which can change during the two hours of a Grand Prix), not to mention changing fuel loads and changing track conditions due to the weather and rubber build up, there are too many variables to cope with.

Perhaps a simpler solution is to build simpler cars and let the ability of the driver make the difference. Oh crap, I am becoming DOF_Power...
 

Grizzly

Bear
Contributor
Hmm, you are also assuming that we actually do understand turbulent flow. From what I have seen, we don't. Therefore any definitions that we have presently are only approximations.

I wouldn't suggest we understand turbulent flow, but i believe we can understand turbulent flow, in years to come. They are approximations because we are still only just scraping the surface with fluid mechanics calculations. We have the computer capacity to make some quite accurate predictions of simple motions in simple scenarios, but I'd agree it will be years before hugely complex calculations like as you say, turbulent flow, can be anything like accurate simulations. That's not to say we can't, we just can't yet!


The narrow rear wing and snowplough front wing that the cars presently sport are direct results of the OWG's work.

Did they put these forward as methods to increase overtaking, or were they asked to come up with new wing regulations?
 
Top Bottom