CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   New to CFD - automotive application question (http://www.cfd-online.com/Forums/main/14590-new-cfd-automotive-application-question.html)

Don January 4, 2008 14:32

New to CFD - automotive application question
 
As a hobby, I do club racing and other driving events. I am to the level and traveling at speeds where aerodynamics are playing a large role in the handling of the car. Most aerodynamic products for car enthusiast on my budget are geared toward 'show cars'. More for looks then actual aero effectiveness. I am in the works of trying to build my own components and would like to start by using a CFD program. Problem is, most automotive CFD programs that I have seem run in the 25,000$ range. I'm not a flipping F1 racer. Is there any CFD programs that are cheap/free that will show wind/aero flow effects?

Bikini Girl with MachineGun January 5, 2008 15:28

Re: New to CFD - automotive application question
 
I know that I'll sound disingenuous, but I truly don't understand why CFD will be helpful to you.

For sake of example, suppose you are aiming to improve your performance by 1-2%. That doesn't sound like much, but it is significant in competitions.

Furthermore, suppose that you have two competing designs and two different gridding schemes for each of them. How will you decide which of the four possibilities is the best given that predictions will probably differ by far more than 1%? (Am I being over-pessimistic and cynical, or can a *complete* vehicle be gridded easily and reliably these days?)

I hope that you don't spend $20k+ for code that probably will do nothing for you. For that sort of money (+ time + effort) you would probably find more than 1% improvement by concentrating on the transmission and other non-aerodynamic factors.

Of course, I'm happy to be convinced otherwise. CFD purveyors and snake-oil salesmen, do your best!

BGwMG


gocarts January 5, 2008 22:46

Re: New to CFD - automotive application question
 
You might consider an affordable panel method, which, while not being as general as field-based CFD, can produce results for streamlined shapes such as closed-wheel race cars. Find out more at http://www.symscape.com/blog/why_use_panel_method

In the interest of full disclosure, I represent Symscape, which produces a Panel Flow add-on (US$99.99/year) for the unified simulation environment SymLab.


Charles January 6, 2008 02:20

Re: New to CFD - automotive application question
 
Most CFD old-timers will advise you that spending money on CFD when competing at your level won't help much. They are probably right. A better idea is to make friends with an aerodynamicist!

Ahmed January 6, 2008 03:31

Re: New to CFD - automotive application question
 
The above two responses have a lot of truth, now, as I understand your post, you are looking for a program to help you design some external accessories, am I right, Here is something for you:- Speeds involved in your application are guaranteed to be subsonic, and definitely the Mach number involved would be less than 0.3, i.e., you do not need to consider the compressibility effects, or using the correct fluid dynamics terminology, your flow can be treated as incompressible flow. Here is my opinion: 1- Download Gmsh, this is a free mesh generator, play with it a bit, if you can construct your geometry with it, then it is fine 2- Down Load Open Flower, another free incompressible solver and is designed to read the mesh generated by gmsh. Search the software link on this forum or google for both. Good Luck

Mayur January 7, 2008 07:29

Re: New to CFD - automotive application question
 
if u have a cad geometry u can try Acusolve... it is much cheaper than other codes and will solve ur purpose.

Don January 8, 2008 13:10

Re: New to CFD - automotive application question
 
The reduction of 'drag' on the car will assist in top end speed. But on most road coarse tracks, the fastest car does not always make the fastest lap times! It is all about corner entry and exist speeds. My concern is more toward the generation of downforce by manipulating the pressure differentials under the car. The more downforce, the more 'grip' the tires will have to the road. Thus I can carry more speed into corners and brake later. There are two areas of interest. The front of the car and the rear. There are some things to notice here:

"http://www.dwhrf.com/don/videos/VIR Top of hill 1.wmv"

If you look at the steering wheel, I keep having to make a lot of minor correction to the steering at I go past 135. This is due to the rear of the car wiggling back and forth on me. If you listen very closely, as I crest the hill, the car's rpm's jump a couple hundred revs. This is due to the front end lifting. Thus I'm loosing traction and loosing acceleration.

Without going into a tremendous amount of detail, I'll cover the front end. The goal is to general 'down force' on the front by using a front splitter.

Here is how it was explained to me:

"A splitter generates downforce on the front of a car by creating pressure differentials. Actually a splitter produces a "downward force". It is not technically termed "downforce" unless the downward force is large enough to overcome the "upward" force caused by lift. But that is really just semantics.

http://e30m3performance.com/myths/sp.../splitter2.gif

The oncoming air approaches the vehicle at V1, which is the speed of the vehicle, and a pressure P1 which is related to V1 according to Bernoulli's equation. P1 will be somewhat less then atmospheric.

Figure 1 shows that as the incoming air reaches the front of the vehicle it must come to a stop before it turns to move either up and over, down and under, or around the vehicle. The area where the oncoming air flow comes to a stop is termed the "stagnation point". Since the velocity has gone down, the pressure has gone up (the "stagnation pressure). Thus the front of the moving vehicle is an area of relatively high pressure.

Thus the relatively high stagnation pressure at the front of a moving vehicle pushes back on the vehicle (creating drag). Now if we stick a splitter out on the bottom of the spoiler, then the stagnation pressure will also push down on the top of the splitter. But what is the pressure on the underside of the splitter? If it is the same as the pressure on top then there will be no force on the splitter. So let's examine the pressure underneath the splitter.

Since the splitter is close to the tarmac, it creates a restriction to air flowing underneath it (a mini-Venturi if you will). Just as water in a river speeds up when the river narrows, the air that is piled up in front of the spoiler must speed up if it wants to squeeze underneath the splitter. And as we know from the Bernoulli equation, when an airstream speeds up, its pressure goes down. Thus the region between the splitter and the pavement is an area of low pressure. To recap; high pressure exists on top of the splitter, and low pressure exists beneath it. This adds up to a downward force on the splitter. And this downward force is directly proportional to the surface area of the splitter. So, up to a point, the bigger the splitter is, the more downforce it can create."

And there is a point of 'too much' downforce, which 'bogs' the car down at high speeds. This is where I think a CFD program would assist in helping me play around with different designs and their effects.

gocarts January 9, 2008 10:04

Rear Diffusers
 
If you are also interested in rear diffusers then you might consider reading: http://www.symscape.com/blog/secrets_of_diffusers

enGits January 10, 2008 08:24

Re: New to CFD - automotive application question
 
Hello,

CFD can certainly be a helpful tool for the kind of problem you are presenting. Be aware, however, that for useful results you need some experience in fluid mechanics, as well as CFD. OpenFOAM (www.opencfd.co.uk/openfoam) should be able to capture the effects you are looking for.

We are a young company, specialised in the field of open-source CFD. What we do is assist customers in the usage of open-source tools. Furthermore we have commenced a sponsored development in the field of open-source mesh generation; as you might have discovered, there is still a lack of open-source mesh generation for complex flow problems.

If you are interested, please feel free to have a look at our website (www.engits.com) and to contact us at info@engits.com. Your kind of problem could be a good example to test and demonstrate the capabilities of the new development.

Oliver



All times are GMT -4. The time now is 10:36.