Hobbyist new to CFD and needs for racing

sm08_jb · January 4, 2012, 16:31

Hi all. I am new to CFD. I have a degree in Industrial Design and seen CFD used in the formula one world so I know of CFD and it capabilities. It has always been interesting to me but I have never really done anything with it. However I now need it.

I race enduro lay down go karts on road racing circuits. The karts have full body work and at Mid Ohio average over 90 mph and hits speeds over 120mph on the straights. Aerodynamics are very important but it is hard to try new stuff because it is just me making the body work and track time is very limited.

Because of my ID knowledge I have built the kart and various body work designs in CAD (rhino 3d). Now I would like to use these CAD models to do CFD analysis of COD and downforce.

My question to everyone is, is there a good software for me to use for not much money. Or is their a trial software I could use to get these numbers. I have used next limit's realflow for water simulation which should give me at least an idea of how to use CFD software. I looked into Next limit's x-flow but can't find much info on how to test it. Or is this crazy to think I could jump into CFD at all because it is so much more complex than using other CAD or simulation programs. In that case...

If I can provide CAD files of my kart would someone be willing to drop in my models and give my data on the different bodies. If so please contact me with what I would need to provide and how much you would charge. Thanks for everyones time.

husker · January 5, 2012, 01:10

Hi Baldi,

It is very surprising to see someone is not a physics-based engineer is interested in CFD.

I think you've heard about OpenFOAM, open-source CFD which is very promising for low-cost solutions to the CFD problems. As I know, it is free.

Martin Hegedus · January 5, 2012, 11:54

I also suggest using OpenFOAM for incompressible flow. (Compressible flow is a different story, IMO, the verdict is not in) But start with simple 2D test cases of your own making. Personally, I don't like their example cases. They are not robust. I have had experiences where I changed the flow conditions on their example cases slightly and the cases failed to converge.

The process of getting a CFD run to converge is not too challenging. It may take a little knob turning and question asking, so patience is required. However, getting a numerically good CFD run can be challenging and take a lot of resources. Also, the pool of people that can help you with this is smaller. So, you must decide if you want qualitative or quantitative results. Good quantitative runs can take at least an order of magnitude more time, etc. than qualitative runs.

From your problem description, it reads as if your Reynolds number may be in the transition region. It will be very difficult to get good quantitative results and if the problem is unsteady, it will take a lot of cpu power.

Good Luck.

January 4, 2012, 16:31	Hobbyist new to CFD and needs for racing	#1
sm08_jb New Member Jeremy Baldi Join Date: Jan 2012 Posts: 1 Rep Power: 0	Hi all. I am new to CFD. I have a degree in Industrial Design and seen CFD used in the formula one world so I know of CFD and it capabilities. It has always been interesting to me but I have never really done anything with it. However I now need it. I race enduro lay down go karts on road racing circuits. The karts have full body work and at Mid Ohio average over 90 mph and hits speeds over 120mph on the straights. Aerodynamics are very important but it is hard to try new stuff because it is just me making the body work and track time is very limited. Because of my ID knowledge I have built the kart and various body work designs in CAD (rhino 3d). Now I would like to use these CAD models to do CFD analysis of COD and downforce. My question to everyone is, is there a good software for me to use for not much money. Or is their a trial software I could use to get these numbers. I have used next limit's realflow for water simulation which should give me at least an idea of how to use CFD software. I looked into Next limit's x-flow but can't find much info on how to test it. Or is this crazy to think I could jump into CFD at all because it is so much more complex than using other CAD or simulation programs. In that case... If I can provide CAD files of my kart would someone be willing to drop in my models and give my data on the different bodies. If so please contact me with what I would need to provide and how much you would charge. Thanks for everyones time.

January 5, 2012, 01:10		#2
husker Member Join Date: Mar 2009 Posts: 85 Rep Power: 17	Hi Baldi, It is very surprising to see someone is not a physics-based engineer is interested in CFD. I think you've heard about OpenFOAM, open-source CFD which is very promising for low-cost solutions to the CFD problems. As I know, it is free.

January 5, 2012, 11:54		#3
Martin Hegedus Senior Member Martin Hegedus Join Date: Feb 2011 Posts: 500 Rep Power: 19	I also suggest using OpenFOAM for incompressible flow. (Compressible flow is a different story, IMO, the verdict is not in) But start with simple 2D test cases of your own making. Personally, I don't like their example cases. They are not robust. I have had experiences where I changed the flow conditions on their example cases slightly and the cases failed to converge. The process of getting a CFD run to converge is not too challenging. It may take a little knob turning and question asking, so patience is required. However, getting a numerically good CFD run can be challenging and take a lot of resources. Also, the pool of people that can help you with this is smaller. So, you must decide if you want qualitative or quantitative results. Good quantitative runs can take at least an order of magnitude more time, etc. than qualitative runs. From your problem description, it reads as if your Reynolds number may be in the transition region. It will be very difficult to get good quantitative results and if the problem is unsteady, it will take a lot of cpu power. Good Luck.