Using SolidWorks Simulation for Submarine 1

[gcohn2]

Hi. I'm trying to run a simulation on a submarine at 1000ft of depth. I set it up using a constant pressure, but I'm having trouble with the 'fixtures'. In reality, nothing will be 'fixed' as it will be completely surrounded by water. I tried using the 'soft springs' option as well as adding anchor points to simulate it being tied down with ropes. I am applying pressure to all of the external surfaces, so I would expect the assembly to stay put, but instead there is a strong resultant force in the y direction, as if there is some other force acting on it. I don't have gravity turned or any other external forces. Has anyone solved a problem like this?
Thanks
Gabe

 

[GregLocock]

Long soft springs should work. I'd be inclined to find the source of the Y force as it sounds like a modelling error - perhaps you haven't applied pressure to /all/ external surfaces.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MintJulep]

Buoyancy.

 

[GregLocock]

" I don't have gravity turned or any other external forces." If there's no gravity there shouldn't be any buoyancy and the hydrostatic pressure should be constant all over the external surfaces.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[gcohn2]

Thanks Greg. Is there a way to adjust the spring variables under 'soft springs' I agree that the first thin to do is figure out this mysterious force. I double checked that all external surfaces are selected. I'll let you know if I figure it out.

 

[GregLocock]

"Is there a way to adjust the spring variables under 'soft springs' "

I haven't the faintest idea. In most mickey mouse programs you either doubleclick on the part in question or right click to bring up a context menu.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[PeterSl]

I have heard of Simulation 2010 randomly loosing applied pressures this could cause the imbalance.
Can you split the submarine on any symmetry planes? You can then effectively prevent movement perpendicular to the plane with the added benefit of reducing the run time.

Regards
Pete

 

[TheTick]

It's finite element analysis, not infinitesimal element analysis. There will always be errors in the pressure distribution. If your model is unconstrained, there will be a net force that causes it to move.

 

[fegenbush]

TheTick is correct and his post is well stated. The solution would be to split your model into halves and constrain the boundary.

 

[gcohn2]

The part is symmetrical about the vertical plane, but not the horizontal and strangely enough I'm getting no net forces in the horizontal direction. Maybe it's a meshing issue. I'm going to try to tighten up the mesh and see if anything changes.

Thanks for all of the input.
Gabe

 

[gcohn2]

Not sure if there is a way to mark this as solved, but it is solved. In case it helps anyone...I had an assembly with a dome mating to a flange. Part of the flange surface was exposed to the water and part of it was not. When I selected the flange surface, I neglected to notice this. To resolve this, I added a split line to the surface, so that I could select only the exposed portion of the flange to apply pressure to. Of course none of you could have figured this out without taking a look at the model and the setup, but maybe one day you will have a similar situation and you can apply a similar solution.

thanks again for your responses,
gabe

 

[GregLocock]

Thanks for posting the solution.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?