[Rehaan]

Originally Posted by Shan2nu

...
Only question with the trailer test is that drag and downforce depend on wind speed and not vehicle speed. A car doing 190kmph with 10kmph tail wind will actually have a wind speed of 180kmph ...

I guess this can be referred to as "air speed" and "ground speed" - similar to aircraft.
Eg. when flying with a constant throttle, a tailwind will increase your ground speed while your airspeed stays roughly the same.

Originally Posted by Shan2nu

...So we would need a sensor that can calculate the exact wind speed hitting the car, irrespective of the trailer speed.

Device to measure wind speed = Anemometer.

There are many types : LINK

Originally Posted by Shan2nu

So, the only options are a wind tunnel capable of 300kmph or CFD model of the car, i guess.

...Or a scale model in a smaller (much more affordable) tunnel as mentioned.

cya
R

PS - Remember in 2004 we said "howcome we don't have dynos like these in India??", and 4 years later we did. Lets see how this goes!

[Shan2nu]

I guess this can be referred to as "air speed" and "ground speed" - similar to aircraft.
Eg. when flying with a constant throttle, a tailwind will increase your ground speed while your airspeed stays roughly the same.

Yup.

...Or a scale model in a smaller (much more affordable) tunnel as mentioned.

But if someone like Jitu wants to test his car with diff body kits. Making a scale model might turn out to be time consuming.

PS - Remember in 2004 we said "howcome we don't have dynos like these in India??", and 4 years later we did. Lets see how this goes!

Exactly......i was thinking on those lines but thought better not to mix wind tunnels and dynos on the same thread. But the idea remains the same.

The wind tunnel in the vid is nowhere as advanced as the ones used by racing teams or even car manufacturers. But for the regular enthusiast or tuner, any knowledge is better than none.

And the Cyber Evo is no pushover, it laps the Fuji speedway in 1min 41secs and the Tsukuba track in 54secs.

Shan2nu

[Udayrg]

It would be much more cheaper and easier to conduct a CFD analysis then a scale/full- model wind tunnel analysis.

Now if we could get a good cad file may be a *.stl , *.iges or *.stp file we can certainly do a CFD Analysis.

I know its tough to get these file however on internet you would find Surface geometries of many popular vehicles.

And then for some modification say rear spoiler we can modify n Number of times to get exact -ve lift at particular speed.

In simple word CFD is very cost effective and has very short TAT (Turn Around Time)

Uday.

[Rehaan]

Originally Posted by Shan2nu

...
But if someone like Jitu wants to test his car with diff body kits. Making a scale model might turn out to be time consuming...

Possible. However, compare that to :

1) Cost of renting a wind tunnel
2) Cost of actually getting all the body kits he wishes to test


Now, alternatively:

This could be done on a smaller scale with a CNC'd model. Might work out cheaper.

OR

Could be done using CFD, tweaked so very easily for optimal results (as opposed to making a new scale model or full size body-kit changes) and then the actual body kit could be made based on the computer model itself.


cya
R

@ Udayrg - How do you do most of your work. Is it all CFD, or does it progress CFD -> Scale testing -> fullsize ?
Also what components take up the most time / are most challenging. Eg. Intake system? Overall aerodynamics? etc

[neel385]

Originally Posted by Rehaan

I guess this can be referred to as "air speed" and "ground speed" - similar to aircraft.
Eg. when flying with a constant throttle, a tailwind will increase your ground speed while your airspeed stays roughly the same.



Device to measure wind speed = Anemometer.

There are many types : LINK
..

Not entirely true. An anemometer is used to measure the speed of the wind as it blows. However, in pressure applications as in the case of calculating the drag on a body, a pitot static device as used in an aircraft will have to be used.

[Shan2nu]

Not entirely true. An anemometer is used to measure the speed of the wind as it blows. However, in pressure applications as in the case of calculating the drag on a body, a pitot static device as used in an aircraft will have to be used.

Rehaan wasn't talking about calculating drag. He was only talking about wind speed or air speed.

Possible. However, compare that to :

1) Cost of renting a wind tunnel
2) Cost of actually getting all the body kits he wishes to test

How much would it cost to get an identical CFD model of say an OHC done, with the various body kits? I remember you were working on a 3d model of a Carrera GT few years ago, were you able to complete it?

Shan2nu

[sgiitk]

There is a 3m wind tunnel in IITK. To my knowledge at least one car firm, and one coach builder have used it.

[Rehaan]

Originally Posted by Shan2nu

...
How much would it cost to get an identical CFD model of say an OHC done, with the various body kits?

No idea what the going rate would be for something like that. Also it would depend on how accurate the model had to be, and how the model was made.

For example - from actual measurements, vs a 3D scan of a 1:18 model.

Perhaps Udayrg can provide some insight (see his "occupation" field).

Originally Posted by Shan2nu

...I remember you were working on a 3d model of a Carrera GT few years ago, were you able to complete it?

I left it at about 85%

cya
R

[Udayrg]

Originally Posted by Rehaan

How do you do most of your work. Is it all CFD, or does it progress CFD -> Scale testing -> fullsize ?
Also what components take up the most time / are most challenging. Eg. Intake system? Overall aerodynamics? etc

Well my work is purely CFD, However my company does some Product development where and when needed i assist in testing and validations.

Most of the automotive domain CFD analysis are some or the other way complex. The complicity increased as physics goes on becoming complex. Like for example a simple flow analysis of a exhaust manifold is quiet simple. whereas a Stored gas air-bag inflators flow analysis would be very complex.

regarding External aerodynamics most of the time is consumed in Clean up, pre-processing (removing insignificant attachments like wipers, Meshing).

Originally Posted by Rehaan

No idea what the going rate would be for something like that. Also it would depend on how accurate the model had to be, and how the model was made.

For example - from actual measurements, vs a 3D scan of a 1:18 model.

Perhaps [/i] [i]Udayrg can provide some insight (see his "occupation" field).

CFD analysis is done on Per hour cost. when all the required inputs are given.

well any CAD model which is not the original will have differences. But again formidable if say the model has 5 to 10 % variation in curvature or surface definitions you can be sure to get say around 80% close results to the original one.

Uday