CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   Motorcycle CFD (http://www.cfd-online.com/Forums/star-ccm/86061-motorcycle-cfd.html)

champion March 13, 2011 17:52

Motorcycle CFD
 
Hi, I am a student currently doing a thesis on motorcycle aerodynamics. I am very new to any CFD software and have not studied much aerodynamics. I have been teaching myself Star-CCM+ from a meshing tutorial which I was given.


Tomorrow I am going to put a scale motorcycle into a small wind tunnel and take a drag reading through a fixture that screws into a load cell in the wind tunnel floor.

My star-ccm+ is using identical: model size and velocitys. This means that the rod which connects between my bike and load cell goes through the floor of the region. (which was set as a symetry plane)

I made the scale model which I imported as an X_T file. I then tried to fix the model representations - import [right] - repair surface (I then deselect the two bottom check faces)

Problem is - it refuses to surface mesh, the error is as follows:

Worst quality is 5.523e-005
initial number of faces: 66732

starting: Apply Global Limits
Error during surface remeshing, deleting resulting mesh
Parasolid function PK_CURVE_eval_curvature failed.
Error token = PK_ERROR_bad_parameter
Error code = 120
Severity = mild
Command: GenerateSurfaceMesh
CompletedCommand: GenerateSurfaceMesh
In: [Machine::main]
Recoverability: Non-recoverable
ServerStack: [SymInit: Symbol-SearchPath: '.', symOptions: 530, UserName: '07034428'
, OS-Version: 6.1.7600 () 0x100-0x1
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 000000000052DE9D)
, 000000000052DE9D (StarPlatform): (filename not available): SingletonCleaner::SingletonCleaner
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000000545919)
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 00000000003BFC2D)
, 00000000003BFC2D (StarNeo): (filename not available): NonRecoverableError::NonRecoverableError
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 000000000038155B)
, 000000000038155B (StarNeo): (filename not available): Controller::executeCommand
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000073466BF0)
, 0000000073466BF0 (MSVCR90): (filename not available): tanhf
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000073433821)
, 0000000073433821 (MSVCR90): (filename not available): _is_exception_typeof
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 00000000773DF661)
, 00000000773DF661 (ntdll): (filename not available): RtlRestoreContext
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000000380E16)
, 0000000000380E16 (StarNeo): (filename not available): Controller::executeCommand
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 000000000037F705)
, 000000000037F705 (StarNeo): (filename not available): Controller::processCommands
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 00000001800122E1)
, 00000001800122E1 (StarMachine): (filename not available): CommandController::processCommands
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000180073421)
, 0000000180073421 (StarMachine): (filename not available): Machine::startServerHost
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000180067006)
, 0000000180067006 (StarMachine): (filename not available): Machine::main
, ERROR: SymGetSymFromAddr64, GetLastError: 487 (Address: 0000000140001297)
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000140001297)
, 0000000140001297 (star-ccm+): (filename not available): (function-name not available)
, ERROR: SymGetSymFromAddr64, GetLastError: 487 (Address: 0000000140004CA2)
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 0000000140004CA2)
, 0000000140004CA2 (star-ccm+): (filename not available): (function-name not available)
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 000000007718F56D)
, 000000007718F56D (kernel32): (filename not available): BaseThreadInitThunk
, ERROR: SymGetLineFromAddr64, GetLastError: 487 (Address: 00000000773C3021)
, 00000000773C3021 (ntdll): (filename not available): RtlUserThreadStart
]
error: Server Error

Any thoughts and advice would be brillaint.

Alex

champion March 13, 2011 17:55

http://i665.photobucket.com/albums/v...on/Capture.png

alastormoody11 March 14, 2011 03:32

Hi,

I havent gone through your case thoroughly but your mesh is showing free edges, which might be the problem.

As along as there are free edges your volume meshing wont work.

You have to get rid of all the extra surfaces, since you are interested in only external aerodynamic study you need to delete all the interior surfaces inside the bike, which I guess are not intersecting the symmetry plane leading to free edges.

Secondly if you imported your geometry in a step format it will take it as an existing surface mesh which I guess is why it is showing quality values before creating the surface, please refer the control valve meshing tutorial for the same, also if you had already created the surface mesh with a quality 10e-5 its very bad so try adding feature curves to enable better mesh generation.

Jonny6001 March 14, 2011 07:29

Maybe the surface wrapper is your best option?

plaunie March 15, 2011 08:22

1. you my want to start back at the cad geometry and check that for errors. I use solidworks and lots of times I can trace my meshing issues to small errors in the cad model. re-import your neutral format into your cad software and use that as a guide to find erorrs in your model. You'll probably end up re-modeling some features.

Also you model is incredibly detailed looking - you may want to start with something simpler. Try running your analysis with cad geometry made of some basic shapes representative of the m-cycle, then add complexity to bring your results closer to what you get out of the wind tunnel.

2. Try different formats to get the mesh into star. I seem to have the best luck with parasolids. (worst luck with step files)

Good luck

champion March 19, 2011 07:50

Thank you for your help guys - Surface wrapper did actually fix the problem...

http://i665.photobucket.com/albums/v...validation.png

champion March 19, 2011 07:57

as you can see the mesh was not perfect around the front forks or inside of the wheel, however it has given me a good set of reasults.

I put an identical size model motorcycle into the wind tunnel yesterday to compare the reasults and got a 0.68N drag from it.

Which is a brillaint reasult. The reason the CFD run has a slightly lower drag is simply the CFD ran with removed handlebars, exhaust, engine, indicators, mirrors, rear suspension .

Thanks for your help.

Here is a photo of the bike in the tunnel. http://i665.photobucket.com/albums/v...n/P1000337.jpg


All times are GMT -4. The time now is 06:43.