External Shell disgn on Solidworks

[Alexisonsmith]

Hey,
I wonded if anyone new how to design the external shell of a FS car on solidworks? Is it easy since there are no tutorials on the student edition.

Regards,

Alexisonsmith

[Lorenzo Pessa]

Are you talking about a not structural cover over a tubular space frame?

[Alexisonsmith]

Yes,

It is too design a glass fibre(I think) external shell to put over a tubular space frame

[t21jj]

Try looking up the Solidworks surfaces tutorial and help files. Our team has done it using surface models the past couple of years but has also modeled it as a solid in the past (I like surface models better). If it's non structural just make sure it follows the rules, can be pulled off the mold, and looks good to you.

[Solidworks FSAE Tutorials]

Hey Alexisonsmith,

Look for a youtube tutorial from me on this in the next couple of weeks. This is a complicated topic so it will probably at least a couple parts

[Alexisonsmith]

Ok thanks http://fsae.com/groupee_common/emoticons/icon_smile.gif,

Could you link me with the youtube URL when it is done,

Regards,

Alexisonsmith

[Alexisonsmith]

Thats great http://fsae.com/groupee_common/emoticons/icon_smile.gif

Could you please post the link when it is done,

Regards,

Alexisonsmith

[Lorenzo Pessa]

some hint

work in an assembly. load the frame and the other stuff that could interfere with your shell.

if you have design sketches (front, top, side) load them in planes far away.

I suggest to start fixing some main points, then trace some spline to connect them then work with surfaces. I think it's the best way to control the shape of your shell.

If you need some advanced surface shaping (and you need to make some CFD) swicht to CATIA (or at least Pro/E).

[Alexisonsmith]

This was going to be my next question, is it possible to simiulate some flows over the surface shape on solidworks? I think not and so would CATIA be the next best thing?

Alexisonsmith

[Adambomb]

That just reminds me of an issue we've been fighting with: Trying to do CFD on the car, and it ends up calculating flow through the inside of the chassis tubes. I remember we had a lot of trouble with it, never quite got it to work right. Although I imagine it could still be done in Solidworks, would just have to perhaps create a shell over the whole assembly as a separate part, then do analysis on just that part.

[t21jj]

Originally posted by Alexisonsmith:
This was going to be my next question, is it possible to simiulate some flows over the surface shape on solidworks? I think not and so would CATIA be the next best thing?

Alexisonsmith

It is if you convert your surface into a solid model.

[Fred G]

Originally posted by Adambomb:
That just reminds me of an issue we've been fighting with: Trying to do CFD on the car, and it ends up calculating flow through the inside of the chassis tubes.

Try building a chassis with round solid bar... the model would weigh a lot, but at least air can't go through the tube http://fsae.com/groupee_common/emoticons/icon_smile.gif

Fred

[Lorenzo Pessa]

If you have to make a CAD you have to use Solidworks, CATIA, PRO-E, ...

If you have to make a CFD you have to use Star-CCM+, Fluent, ... In this case, remember to avoid Solidworks to model the CAD because you will get crazy after import it in a CFD program.
Last year we used Solidworks. Importing the simplified CAD in StarCCM+ results in too many hundreds of errors.
So we decided to switch to CATIA.

[Alexisonsmith]

So I would just do the CFD in Solidworks? Or is there something better which I can just transfer it to?

Alexisonsmith

[t21jj]

Originally posted by Alexisonsmith:
So I would just do the CFD in Solidworks? Or is there something better which I can just transfer it to?

Alexisonsmith

You can do CFD in Solidworks it's not the best program but it's not hard to run. You would just have to convert the surface model to a solid. I've not been able to make a surface model run in Flow Simulation yet.

[Alexisonsmith]

would it not better to transfer the model to another program such as Star-CD or Fluent?

[t21jj]

Ultimately yes they would be better, though it depends on what your goals are and how much time you want to spend.

[Alexisonsmith]

Is Star CD easy to use?

[t21jj]

It's got a learning curve.

If you're that serious about it I'd find someone at your uni who can show you the basics. Or might know a different program.

[Solidworks FSAE Tutorials]

Check out the SolidWorks built in tool, SolidWorks Flow. It is an addin, so make sure to activate it by going to the Tools Menu and selecting Add Ins and then activating it.

To do flow simulation you will have to convert to a solid model. You won't get accurate results using just a surface model.

If you have any questions or need help shoot me an email: sfaulkner at solidworks.com

[Lorenzo Pessa]

Originally posted by Alexisonsmith:
Is Star CD easy to use?

It's better to use Star-CCM+ than Star-CD, it's a more powerful environment to perform CFD simulation.

the CFD user interfaces are developing to "user friendly" but aren't so friendly yet.

There are few tutorials in the guide of Star-CCM+. With patience and a little time you will be able to work with the program.

The problem is setting up the correct parameters for a car CFD simulation.

[Solidworks FSAE Tutorials]

Hey Everyone,

Here is the link to the surfacing tutorial part 1. There will be a second part coming.

http://www.youtube.com/watch?v=-8dKweEt1OQ

[Alexisonsmith]

Thanks alot this is just what I need

[Adam Vaughan]

I will say that this is exactly what my CFD professor drove home my first day of class: The acceptable tolerances for CAD are much lower than those for CFD.

BUT:

I've had very good luck going from a fairly complicated SolidWorks fluid volume -> Export ACIS (ver. 16) file format -> import into GAMBIT (with geometry heal) -> Tet or Hexcore mesh -> FLUENT. Any issues I had meshing were due to small faces that you could design around in SolidWorks with a healthy dose of extrudes/fillets/move faces that didn't appreciably change the shape of the fluid volume. GAMBIT has the ability to show the most skewed cell, which helps in finding these little faces.

The reason I prefer this approach is that GAMBIT is a terrible program to draw anything more complicated than a solid block/stick figure. It IS, however, a fairly powerful meshing program... especially when you use size functions. Also, while a Tet mesh is not ideal... you'll very quickly see why it saves human time when meshing complicated shapes. Use the Hexcore if you want something "more accurate." There are other meshing programs, such as HyperMesh that might be worth looking into... I just haven't had a chance to mess with them.

If your school has a CFD course, I'd strongly recommend taking it, as there are a lot of little nuances to doing CFD properly. And no offense to FloWorks, but FLUENT is a much more powerful program to work with. I haven't used Star-CD or Star-CCM+.


Originally posted by Lorenzo Pessa:
If you have to make a CAD you have to use Solidworks, CATIA, PRO-E, ...

If you have to make a CFD you have to use Star-CCM+, Fluent, ... In this case, remember to avoid Solidworks to model the CAD because you will get crazy after import it in a CFD program.
Last year we used Solidworks. Importing the simplified CAD in StarCCM+ results in too many hundreds of errors.
So we decided to switch to CATIA.

[Adam Vaughan]

As a complicated (but non-FSAE) example of what can be done with the SolidWorks to Fluent approach, here is an oxygen uptake study I did a few months ago for a professor growing bone cells on a bio-scaffold:

http://notcooperedu.org/~vaugh...CFD%20Simulation.pdf (http://notcooperedu.org/%7Evaugha/Original%20UDF%20Oxygen%20Transport%20in%203d%20Sc affold%20CFD%20Simulation.pdf)

[Solidworks FSAE Tutorials]

For everyone who was waiting for part 2 of the surfacing tutorial. Here it is:

http://www.youtube.com/watch?v=-MFFjb6ul4A

[Alexisonsmith]

Thank you very much http://fsae.com/groupee_common/emoticons/icon_smile.gif