CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   LOOKING 3D tutorial and guidance (https://www.cfd-online.com/Forums/fluent/97551-looking-3d-tutorial-guidance.html)

 banun February 20, 2012 08:55

LOOKING 3D tutorial and guidance

Hello..

I am FLUENT beginner..i have thesis to do for my degree and the analysis is using FLUENT.
my thesis is about an aerodynamics on solar race car.

any suggestion for me to refer any 3D analysis tutorial that analyzed drag and lift coeeficient?

im really need help

TQ

 diamondx February 20, 2012 13:14

Did you try tutorial in ansys Website customer portal ??

 banun February 21, 2012 05:57

..the website need account number (customer number)..

and unfortunately i didnt have it..

my progress, im draw 2 part in Solid Work and assembly it (since im more familiar with Solid Work)
the two part is a car model and a big hollow rectangle as the wind tunnel
after i mate the two part, i save in IGS to be open in ANSYS

now i can proceed for meshing right?.
do we mesh on both part?..i mean the car and the model?

thank you very much for any help...

 banun February 21, 2012 09:34

how to do mesh on 3D model with wind tunnel??

really appreciate for any explanation..

 diamondx February 21, 2012 12:39

I've never done that in SOLIDWORK. But here are the basic step for such an analysis.
What you need to that is to draw your car. then you need draw a big rectangle -your car needs to be within the rectangle- IT IS VERY IMPORTANT TO EXTRUDE YOUR RECTANGLE WITH THE "ADD FROZEN" METHOD, by doing this you can then do a "BOOLEAN OPERATION" on the bodies. this will allow you to delete the car and leave the rectangle with the hollow in it. After that you can proceed to mesh your rectangle in ansys meshing or ICEM-CFD. May be you are not familiar with Ansys desingmodeler to do that. Just give it a try it's an easy tool.
Here's is a video of analysis of a blunt body get some inspiration from here --->

 banun February 22, 2012 01:33

thanks for your explanation and the link diamondx!..=) they give me clear view where i should start and how, regarding to my thesis ...will study it..i hope u dnt mind if asking for any doubt next..

TQ

 kdrbrk April 3, 2012 04:53

Hi TQ,

Actually there are thesis about this subject, but all of them are confidential for the teams. So it is almost impossible to find a sample.

What have you done so far?
3D design, meshing, analysis...?

 banun April 3, 2012 09:55

Quote:
 Originally Posted by kdrbrk (Post 352852) Hi TQ, Actually there are thesis about this subject, but all of them are confidential for the teams. So it is almost impossible to find a sample. What have you done so far? 3D design, meshing, analysis...?

i have done the simulation..but there is some problem..because the results are unreasonable..extremely too small

we get the drag coefficient directly value from the RESULT-REPORTS-FORCES....right?

i think i problem with my scale of my drawing and the velocity

the actual model size is length 5 meter, height 1.6 meter, width 1.8 meter
i draw the model with small scale drawing
theoretically, with that actual size..the Reynolds Number is 7.9 X10^6 and the velocity 55mph

for small scale drawing, i should change the velocity right?..it cant be 55mph anymore..

i have problem to know the correct velocity value for small scale drawing

can u explain to me how to get the correct velocity value for small scale drawing?

is it just use general ratio??..or similitude of reynolds number?..i have try both but the result is still the same...or

mybe can u explain to me the correct step..(confused actually)

MANY THANKS FOR ANY HELP..:)

 kdrbrk April 3, 2012 10:10

Hey,

You don't need to scale your model, just use real values.
So that results can be more accurate.

Use reference values section to set parameters and see coefficients, or just see the force* and calculate the coefficient yourself.

*that is why using real scale is important. you see the actual force.

note that if you use half of the car and use symmetry plane, then it is the half of the force, multiply the force by 2.

hope this helps.

note: was importing model from some other CAD, and it was modeled in milimeters. But fluent realised as meters.

so a solar car with 5000mm(5m) length become 5000m (5km) long. I was using mesh->scale to correct this...

 banun April 4, 2012 01:15

yea..i plan to draw in actual scale..

half body and do symmetry plane...about that force that that need to multiply 2..if u not tell me, i will miss that thing....=)

hope no problem with the computer memory and not take more than a night to run for one simulation.... ;D

about fluent realised as meters, did u set meter when the design modeler launched?
means it is ok to draw in mm in other CAD (solid work mybe)?..since my drawing in mm draw in solid work...

TQ

 kdrbrk April 4, 2012 06:05

I was using CATIA for 3D modeling, and length unit was milimeters, Gambit for meshing, and Fluent for analyzing and postprocessing.

I realized Fluent took my geometry and mesh in meters when I used "check mesh"(or something like that) and saw that domain is kilometers long and wide and high. Later I realized that (after some analysis and some abnormal results) I had to scale my mesh.

A quick tip for you:
If you want to simulate transition using transition models, you need to mesh boundary layer so that y+ does not exceed 1. I could achieve this by setting first node 0,003 milimeters (or 0,03mm can not remember now, it has been a long time) from the surface. That increases the total mesh count significantly.

 All times are GMT -4. The time now is 23:33.