Lotus T127 Cosworth

[Mystery Steve]

can you simulate real flow very precisely?

Generally, yes. The Reynolds number is used to scale the measurements taken on the wind tunnel model to estimate the forces that would take place on the full size vehicle.

Can you simulate car-track interaction?

Some wind tunnels do have "rolling roads" that account for the ground effect.

its behavior under yaw or pitch, influence of the flexibility of its components (wings, tires, suspension, steps and gaps...) ?

Normally with most test models you would assume rigidity of the components. You can get into things like aeroelasticity effects when you have a more or less final design. With a model in the design state shown, you would set a height, pitch, wing setting, velocity, etc and see what the resulting forces are and repeat the process while varying one or more of the settings. Which relates to the next question...

can you measure influence of these fine details with high precision?

It isn't necessarily the absolute value they're chasing. They are looking more at trends: how the wing angle affect wheel loading, sensitivity to pitch/yaw, etc. Yes, the size of a bolt will cause a slightly different drag number if it isn't perfectly to scale, but looking at the big picture it doesn't affect the information they are trying to obtain.

[noname]

(...) to estimate (...)
(...) you would assume (...)
It isn't necessarily the absolute value they're chasing. They are looking more at trends: how the wing angle affect wheel loading, sensitivity to pitch/yaw, etc.

that was exactly my point. it does not make much sense to spend much time (and resources... money in general) to produce exact replica of the car if we can not simulate, not model, conditions in which it would work.

trends and patterns is all we can get from simulations.

regards

[benjabulle]

I don't knwo if it had been already written here but Manor car, designed by N.Wirth company, will be only studied by CFD, no wind tunel models or full scale. I think that's the first time for a F1 car, their last LMP1 Acura was designed the same way.

[Giblet]

I love the folks who can eyeball a car and tell if a team's next season is going to be successful or not.

The model doesn't need to be representative of anything race-able, especially if it's a first pass. I'd rather have a simple wind tunnel model (like this one) in my first go-around to validate my CFD tools before going off the deep end in design.

I have a request, from any modelling/CFD expert on here.

Please make a 3d model of an armchair, and run a CFD on it so I can have a picture of Armchair Aerodynamics.

Seriously, I want it. Would be very funny as an avatar.

[Richard]

I don't knwo if it had been already written here but Manor car, designed by N.Wirth company, will be only studied by CFD, no wind tunel models or full scale. I think that's the first time for a F1 car, their last LMP1 Acura was designed the same way.

That'll be interesting. In my field of tall buildings, a wind tunnel gives far better understanding of behaviour than we can get with CFD. We do use CFD because it can do many models a week, whereas its only realistic to do a couple of tunnel models for the whole project. However, we rely on the tunnel to verify the fundamentals, while the CFD helps us play with design iterations.

The problem with CFD is to do with the wind/structure interaction and aero-elastic modelling. Now I'm talking about relatively stiff static object with relatively simple stiffness characteristics, compared to a F1 car. Will be interesting to see how a pure CFD car performs in reality.

[Mystery Steve]

(...) to estimate (...)
(...) you would assume (...)
It isn't necessarily the absolute value they're chasing. They are looking more at trends: how the wing angle affect wheel loading, sensitivity to pitch/yaw, etc.

that was exactly my point. it does not make much sense to spend much time (and resources... money in general) to produce exact replica of the car if we can not simulate, not model, conditions in which it would work.

trends and patterns is all we can get from simulations.

regards

Engineering is nothing but trends and patterns by its very nature. And I should clarify what I meant by absolute value. No, it is not the exact value that would be experienced on track, but it is more than close enough for practical purposes more often than not. Is it always perfect? No, it's engineering...

And CFD is... engineering. It isn't perfect, either. Like richard said, it is useful for doing design iterations, but the data is dubious if you don't verify it in the tunnel. It absolutely makes sense to produce the model to verify the CFD data.

Again, design is as much about the iterative process as the final product. If I was at Lotus, purchasing this model would make sense to me because it would function as an exercise between the wind tunnel engineers and the CFD engineers.

I second the armchair aero idea. I'll have to look into my 3D CFD access at school and see if I can come up with something.

I would be interested to hear the qualifications of people when they make comments on this website. I don't mean to sound elitist, but how many of the people commenting on this thread have actually used a wind tunnel before? What about design with CFD?

[Jersey Tom]

And CFD is... engineering. It isn't perfect, either. Like richard said, it is useful for doing design iterations, but the data is dubious if you don't verify it in the tunnel. It absolutely makes sense to produce the model to verify the CFD data.

This is a very good point that people should pay attention to.

I think a lot of people have the misconception that CFD, FEA, tire data, and vehicle simulation are all super accurate. They aren't. It's all incredibly sensitive to how its set up. Can be crap just as much as it can be gold.

Everything needs to be validated. Simple models make it much easier.

[LotusF1]

Applause...the only way that you guys think it might be a FI replica is by theory, otherwise axle you work for FI and have been working on next years car and maybe they look very similar...imo its totally different...

You seem to be taking this all personally...

You haven't given any informed counter arguments.
We have a number of facts;
1. Aerolab have a Force India 50% model
2. Aerolab have been used by MG for most of his working career.
3. There are many parts that are IDENTICAL to the VJM02. Floor, monocoque, Front Wing mount points for the flow conditioners, forward sidepod shape, the list goes on.

Answer me this, why would LF1 copy a VJM02? Last years car? Not even at the front of the grid?

Occum's razor.

They haven't, they have borrowed (lol) FI's car.

how can aerolab have a 2010 FI model? I dont understand point 1. For point 2 what do you mean? Still dont see any relevance. Floor is basically written by rules...sidepod??? they are completely different...I can maybe believe that they were inspired by FI, but not the SAME car...common...those are massive accusations and ytou seem sooooo sure, which means you either work at FondTech or FI. Please tell us who?

[czt]

how can aerolab have a 2010 FI model? I dont understand point 1. For point 2 what do you mean? Still dont see any relevance. Floor is basically written by rules...sidepod??? they are completely different...I can maybe believe that they were inspired by FI, but not the SAME car...common...those are massive accusations and ytou seem sooooo sure, which means you either work at FondTech or FI. Please tell us who?

Aerolab were doing work for FI (and previously as Spyker) before MG hired them for Lotus - therefore they must have been using a FI model. The uncanny similarity of the Lotus front wing to that of the FI would tend to back the theory up...

[gridwalker]

I love the folks who can eyeball a car and tell if a team's next season is going to be successful or not.

The model doesn't need to be representative of anything race-able, especially if it's a first pass. I'd rather have a simple wind tunnel model (like this one) in my first go-around to validate my CFD tools before going off the deep end in design.

I have a request, from any modelling/CFD expert on here.

Please make a 3d model of an armchair, and run a CFD on it so I can have a picture of Armchair Aerodynamics.

Seriously, I want it. Would be very funny as an avatar.

+1
High res images please!!!
I have an A2 printer & that would make a brilliant "motivational" poster for my study

[marcush.]

And CFD is... engineering. It isn't perfect, either. Like richard said, it is useful for doing design iterations, but the data is dubious if you don't verify it in the tunnel. It absolutely makes sense to produce the model to verify the CFD data.

This is a very good point that people should pay attention to.

I think a lot of people have the misconception that CFD, FEA, tire data, and vehicle simulation are all super accurate. They aren't. It's all incredibly sensitive to how its set up. Can be crap just as much as it can be gold.

Everything needs to be validated. Simple models make it much easier.

I´d like to make a point here .I´m working in Automotive business and happen to
work in an area where aero is also a factor .So of course there are CFD and Aerodynamic testing in the tunnel day after day.
Is there correlation to what we see in the actual car? yes ,air is flowing ...,but
really I do have massive doubts the results we get are really of big value.
You don´t get real data you could rely on. Everytime some sort of fudge factor is
needed to match the numbers with real world.The guys get very slippery when you try to nail them down on their findings and at the beginning of the project theres no calculated data it is only a scaled data from previous car.
So in my view the whole story is really reducing to just how experienced and all encompassing the dicisiontakers are .You may iterate towards infinity with CFD if you just miss one factor all your results are worthless or may even lead into directions contracticting actual needs.
You don´t need to correlate CFD to tunnel ,both only simulate real world.You need to correlate both to the actual car!Only then you are sure you are not developing rubbish .Stopwatch does not lie.
As my personal experience in Racing I do know quite well how to get tyres in their
operating window and keep them there I make allowances for all kind of stuff in preparation for the next set to be used.Most of the time nowadays I´m spot on with my preparations but honestly speaking I have no clue how to put all this into a formula so some unexperienced newbie stands any chance to make this work at the track .There are so many factors to consider ,with the driver and driver confidence level just two impossible to quantify that I have doubts how this could possibly fit into a excel sheet producing sensible results...

[noname]

Is there correlation to what we see in the actual car? yes ,air is flowing ...
You don´t need to correlate CFD to tunnel ,both only simulate real world.You need to correlate both to the actual car!

http://www.marczyk.net/Beyond%20Optimization.htm
http://www.marczyk.net/downloads.htm

sorry for this off-topic but if we already started discussing numerical methods I've decided to add my penny. I think guy who wrote articles hidden behind links above has quite interesting view on CAE-industry.

regards

[axle]

Applause...the only way that you guys think it might be a FI replica is by theory, otherwise axle you work for FI and have been working on next years car and maybe they look very similar...imo its totally different...

You seem to be taking this all personally...

You haven't given any informed counter arguments.
We have a number of facts;
1. Aerolab have a Force India 50% model
2. Aerolab have been used by MG for most of his working career.
3. There are many parts that are IDENTICAL to the VJM02. Floor, monocoque, Front Wing mount points for the flow conditioners, forward sidepod shape, the list goes on.

Answer me this, why would LF1 copy a VJM02? Last years car? Not even at the front of the grid?

Occum's razor.

They haven't, they have borrowed (lol) FI's car.

how can aerolab have a 2010 FI model? I dont understand point 1. For point 2 what do you mean? Still dont see any relevance. Floor is basically written by rules...sidepod??? they are completely different...I can maybe believe that they were inspired by FI, but not the SAME car...common...those are massive accusations and ytou seem sooooo sure, which means you either work at FondTech or FI. Please tell us who?

Clearly you have the Lotus blinkers on;
It's not a 2010 model, it's a 2009. Aerolab have it as they are contracted to work for FI.
I personally don't work for either. The evidence is there, you choose to ignore it as you are blinkered by bias.

Personally I would have been much happier if LF1 hadn't tried to palm this off as their own in a pathetic attempt to gain credibility.

[autogyro]

I dont see the point of half scale wind tunnels anymore.
Nick Wirth believes the work can be done on computer based on data base.
The final applications are decided by committee meetings anyway.
I suppose it gives something tangible for the boys to play with.
The difference between half and full has to be re calculated from a computer data base anyway, so I do not see it as being development work, just variations on a theme, like most of F1.

[axle]

I dont see the point of half scale wind tunnels anymore.
Nick Wirth believes the work can be done on computer based on data base.
The final applications are decided by committee meetings anyway.
I suppose it gives something tangible for the boys to play with.
The difference between half and full has to be re calculated from a computer data base anyway, so I do not see it as being development work, just variations on a theme, like most of F1.

I believe it's mainly for verification of data. And with limited track testing you want to know the parts (in theory) will do the job.