Mixed Mesh FEA Help

[krederone]

So i have this task of validating our Frame through CosmosWorks, we ended up designing a differential box made of aluminum for the rear.

during analysis our model meshes, using a mixed mesh, but does not run it gives the error of "Restraints may not be adequate." Restraints are in the cylindrical portion of the diff box where the suspension mounts.

Has anyone else ran into this problem.

i will take some screenshots, so you guys can take look at the frame.

also if theres anything that looks that might not meet the rules, please let me know.

http://img.photobucket.com/albums/v301/krederone/screenshot_1.jpg

Thanks!
-Will

[krederone]

So i have this task of validating our Frame through CosmosWorks, we ended up designing a differential box made of aluminum for the rear.

during analysis our model meshes, using a mixed mesh, but does not run it gives the error of "Restraints may not be adequate." Restraints are in the cylindrical portion of the diff box where the suspension mounts.

Has anyone else ran into this problem.

i will take some screenshots, so you guys can take look at the frame.

also if theres anything that looks that might not meet the rules, please let me know.

http://img.photobucket.com/albums/v301/krederone/screenshot_1.jpg

Thanks!
-Will

[The AFX Master]

Dude, taking apart any FEA, that rear end design wont pass tech for sure. You' feeding suspension loads through the engine and through the braces in that way. since there aren't loadpaths to the Main Hoop other than the braces and the engine itself for the suspension loads. That box must be fully linked with the main hoop restraining all the 6 DOF without considering braces AND engine in that procedure, so delete the engine and the braces and join the thing to the Main Hoop.. Then you can add bracing and engine to the design. Be sure that your bracing goes directly to that structure (the structure between the box and the main hoop)

Hope it helps

[Steve Yao]

Is there a 2009 rule that bars fully stressed engines? FSAE teams have been using the engine as a semi-stressed or fully stressed chassis member since almost the beginning of the competition.

Krederone,

You should be have something in there to take the place of the engine as well. ideally, you should also have your suspension package as well. Afterall, it will be the combined stiffness of every component from tire patch to tire patch that will affect your handling and response.

As for your restraints problem, a few things off the top of my head:

1) check your mesh to see that your beams are actually connected to your box solid mesh (or shell mesh)

2) If all you have is cylindrical restraints at suspension pickups, then your chassis has nothing to restrain it from translating longitudinally.

3) Solidworks/Cosmosworks version. I might be mistaken, but I had thought that Cosmos didn't support mixed mesh studies until the 2009 release of Simulation (the "CosmosWorks" name is being replaced with Solidworks Simulation)

As for legality:
Looks technically legit... assuming the members connected to the top of the front roll bar are serving double duty as roll bar support and front bulkhead support.

I would personally prefer better triangulation to major loading points. Your front end looks to have very little to resist lateral bending.

[The AFX Master]

As per 2009 rules:

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">3.12.7 The attachment of the Main Hoop braces must not compromise the function of the bracing
i.e. the attachment method and supporting structure must be capable of transmitting all
loads from the Main Hoop into the Major Structure of the Frame without failing. The braces
must transmit this load directly through a properly triangulated structure back to the bottom
of the Main Hoop. Bracing loads must not be fed solely into the engine, transmission or
differential, i.e. the bracing must terminate at a node where there is a load path back to the
Main Hoop.
3.12.8 If any item which is outside the envelope of the Primary Structure is attached to the Main
Hoop braces, then additional bracing must be added to prevent bending loads in the braces
in any rollover attitude. </div></BLOCKQUOTE>

[scott_rfr]

Couple things, Yes that will not pass tech. If you want check out Rutgers 06-08 cars as we had a very similar design to what your trying to do. Might give you a better idea of load paths. Also something to think about is whats going to happen when you put the quick jack on the back of the car with the roll hoop bracing like you have it. In the end your really going to want a hoop brace going to the forward upper point on that rear sub you have.

As far as the fea goes how do you have your current restraints defined? You want to make sure that you don't under define or over define the constraints. In Ansys you can use a mass block for the engine. Don't know if there's anything like that Cosmos if not just put some al bars across to represent the engine.

Scott

[krederone]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by The AFX Master:
Dude, taking apart any FEA, that rear end design wont pass tech for sure. You' feeding suspension loads through the engine and through the braces in that way. since there aren't loadpaths to the Main Hoop other than the braces and the engine itself for the suspension loads. That box must be fully linked with the main hoop restraining all the 6 DOF without considering braces AND engine in that procedure, so delete the engine and the braces and join the thing to the Main Hoop.. Then you can add bracing and engine to the design. Be sure that your bracing goes directly to that structure (the structure between the box and the main hoop)

Hope it helps </div></BLOCKQUOTE>

Thanks, i appreciate your input.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">1) check your mesh to see that your beams are actually connected to your box solid mesh (or shell mesh)

2) If all you have is cylindrical restraints at suspension pickups, then your chassis has nothing to restrain it from translating longitudinally.

3) Solidworks/Cosmosworks version. I might be mistaken, but I had thought that Cosmos didn't support mixed mesh studies until the 2009 release of Simulation (the "CosmosWorks" name is being replaced with Solidworks Simulation) </div></BLOCKQUOTE>

Thanks, first im going to redesign the rear portion, and take your advice in consideration.

[krederone]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by scott_rfr:
Couple things, Yes that will not pass tech. If you want check out Rutgers 06-08 cars as we had a very similar design to what your trying to do. Might give you a better idea of load paths. Also something to think about is whats going to happen when you put the quick jack on the back of the car with the roll hoop bracing like you have it. In the end your really going to want a hoop brace going to the forward upper point on that rear sub you have.

As far as the fea goes how do you have your current restraints defined? You want to make sure that you don't under define or over define the constraints. In Ansys you can use a mass block for the engine. Don't know if there's anything like that Cosmos if not just put some al bars across to represent the engine.

Scott </div></BLOCKQUOTE>

I just checked out your teams ride, sweet! It did give me better ideas of how to redesign our frame.

I will work on the model this week, and post the progress.

& i did realized i have to input some aluminum bars to act as if it was the engine, since it is a structural member.

[Wes Burk]

The rule about your Main Roll Hoop Braces having a load path traingulated driectly back to the bottom of the Min Roll Hoop os the main issue that I see here. Our car is similar in that it doens't have very much structure rear of the Main Roll Hoop.

C:\Documents and Settings\Wesley Burkman\My Documents\My Pictures\fsae\untitled
We just went through 3 rules clarifications and in the end they are very firm about it going directly to the bottom of the hoop. If it is not too late you should also consider using forward mounted Main Roll Hoop Braces, as you would not have to alter much in the rear.

Have you made a SolidWorks part to represent the clearance templates? It looks like your section between the Main Roll Hoop and the Front Roll Hoop is pretty short.