Cosmos remote load error

[Nihal]

I am trying to use SW COSMOS to FEA an upright design I have and I am having trouble getting the remote loads to work. It keeps giving me the error "Invalid load option. Use FFEPlus." Eventhough I am using FFEPlus. I have been getting this issue alot whenever i use remote load. This happens on both load options, Rigid Connection and Direct Transfer. What am I not understanding about the program/FEA?
Thank You

[Nihal]

I am trying to use SW COSMOS to FEA an upright design I have and I am having trouble getting the remote loads to work. It keeps giving me the error "Invalid load option. Use FFEPlus." Eventhough I am using FFEPlus. I have been getting this issue alot whenever i use remote load. This happens on both load options, Rigid Connection and Direct Transfer. What am I not understanding about the program/FEA?
Thank You

[Chris Boon]

Hey mate,
I used COSMOS to model our uprights and remote loads worked fine for me. All I did was set the restraints, and then defined the appropriate loads. For the wheel bump I selected remote load with rigid connection.

For face I selected the cylindrical section where the hub is mounted and then entered the appropriate coordinates for the remote location and then applied the force. Are you using COSMOS or COSMOSXPRESS?

If you doing all this which I am guessing you are I really cant see whats going wrong cause I have never had an issue with any of the load scenarios.

Maybe you could try to install Solidworks again. Otherwise, you might have to have a chat to a person from the company who can fix your problem.

The only problem I have faced with Solidworks is the alterations in results as the mesh is refined, but i guess when you think about it, it makes sense.

Oh, and the fact that I cant figure out how you can bring assemblies into other assemblies and stop all the components becoming fixed, instead of keeping the previous mates.

Cheers,
Chris

[Nihal]

I am using the regular Cosmos package, not CosmosExpress. I have had this issue with remote loads when I was FEAing our hub design as well. I ha to create a dummy lever arm and apply the loads to that to get it to work, but I never really trusted my results because of all the errors I get. I wonder if Cosmos has issues with porly done CADs, the upright has some complex geometrey and some refrences to dimensions/edges that no longer exist etc.

I have seen differences in stress and defelction levels when I do refine the mesh. I am unsure though how much I need to refine a mesh in order to have a satisfactory answer. I know FEA can only get me to within maybe 25% of the real deflection and stress levels, but I still need to get that close.

Is anyone else having trouble registering with www.eng-tips.com? (http://www.eng-tips.com?) I have tried to register there for a while but it always fails and the support there never responds to emails. They have a large Cosmos discussion forum as well.

[BeaverGuy]

For any FEA or CFD analaysis when you refine the mesh you refine it until the difference between your current analysis and the previous analysis is small enough that you are satisfied that any further refinement in the mesh will not produce better results.

I haven't tried to register with eng-tips lately. However, I did had 2 accounts terminated about 3 years ago. I imagine it was because I was a student though I didn't mention it, i accessed the site from a school computer. Even if you aren't explicity violating their terms of service regarding asking questions about homework they still don't like students posting on the site.

[Chris Boon]

Having poorly done CAD will bugger it up. Complex geometry shouldnt be too much of a problem, because you arent importing the geometry from another completing different program (we have that problem when at uni when we CAD stuff in Solid Edge then do FEA with Strand 7). If the CAD was too complex it would also come up with errors when you mesh and you wouldnt be able to do any modeling for any load case. Using a smaller mesh size usually helps to get around that.

I reckon having references that no longer exist would be a big problem cause when you are doing modeling, the program is still looking to fulfil these specifications when it works (I think). If there are references that no longer exist it will find it pretty hard to fulfil the constraints in certain scenarios, such as remote loads that apply a load as well as a moment.

Cheers,
Chris

[Nihal]

I figured out the issue. I was applying the remote load to what essentially was two faces of the bearing bore (it is a two piece upright). I made a thin (about .01") cylindrical extrusion over the bearing ore and merged the two parts. I no longer got the error after that.

I also figured out the eng-tips thing. It didn't like my berkeley.edu e-mail address. Kind of annoying if you ask me.

[kwancho]

I'm going through the same thing right now, and I have a question about restraints. I put a remote load on the bearing ID, and restrained the balljoints, and I got loads comparable to what I'd done by calculating the individual loads by hand, and pulling on the balljoints with express. But, how can I find deflection at the balljoints? Should I restrain the hub and apply the same remote load at the ballljoints?

[Nihal]

I would look at the deflection of the bearing bore, the camber and toe change.

[Joe Wilkie - SolidWorks]

Hard to say what would cause a remote load error without seeing the model, so I can't be of much help there.


Chris, I can answer your question. Sub-assemblies can be RIGID or FLEXIBLE. You wouldn't want SolidWorks to be concerned with solving mates of a subassembly of say, your frame, gussets, etc, once you've designed it. The parts would be mated so they don't move relative to each other, so it would make sense to treat them as RIGID.
Your suspension, however, would be something where movement is critical to check clearances, etc. It should be FLEXIBLE.

To make a sub-assembly FLEXIBLE, right-click on the subassembly icon in the Assembly Manager, choose PROPERTIES, and check the FLEXIBLE button.

[kwancho]

I've been having a lot of problems getting my pushrod to work in the rear, which is at a complex angle. I can't just mate all the points together, and I tried maiting the pushrod assembly first to a reference plane at the angle, and then to the points, but SWX still craps itself. I also tried mating to a fixed point (ride height), rather than the point on the lower A-arm, which moves. Any suggestions?