Learning Forum

[fabrizio.pirozzi]

Hi guys,

I would like to know how is possible to do this type of analysis in Ansys Workbench:

I have an assembly; starting from a initial position the system moves with a velocity V, reaches the final position and stops. When the system stops, due of its inertia and elasticity it will continue to oscillate for a while and then will definitively stop for damping. I would like to simulate this situation and have an idea of the time to stabilize the system.

I think this simulation is quite ambitious and complex, so I would like some help on which ansys environment to use and what results can be obtained (and if this is possible).

 

Thanks a lot.

Fabrizio.

[jj77]

Sure, just use transient dynamic, assign a displacement vs time on the boundary that is to be moved, and solve it. That should give what you want.

In order to estimate displacement you have the relation a=dv/dt, and v=dx/dt.

 

So say you have a speed of 1 m/s when moving 1 m, then the time duration should be 1 second (for that step - that is the time it takes to move from 0 to 1 m), and then this displacement will stay there (1 m) for the rest of the time which will result/be a free oscillation that will have an amplitude that will be reduced by damping 

 

[peteroznewman]

I agree with jj77 that Transient Structural is the right analysis system for flexible bodies, which I assume is what you have. It's always helpful to insert an image of the assembly.

Only if the assembly has joints and springs that connect practically rigid links would you use Rigid Dynamics.

If the assembly is in an initial position and rapidly accelerates up to a velocity V, the assembly will begin oscillating while it travels at that velocity, and the oscillations will reduce due to damping. 

Do you want to model the oscillations caused by accelerating from the initial position to the velocity V?

If the answer is no, then you can assign the velocity V as an Initial Condition to all the moving parts and that will represent the assembly moving after all the starting oscillations have died out. This is a simplifying assumption and you can travel a tiny distance before applying the brakes.

If the assembly is lightly damped and the time to move to the final position is short, then it is probably still oscillating when the deceleration begins. This is where it gets more complicated. The figure below shows a vertical beam anchored to a frame moving at velocity V. The blue line at the tip of the cantilever shows the motion of the tip as seen from the moving frame. Due to the start up acceleration, the top of the cantilever is oscillating between the extreme displacements 2 and 4 measured relative to the moving frame.

You will get very different responses after you stop the moving frame depending on where the tip of the assembly is at the instant the moving frame stops. For example if the tip is at state 1 and moving to the right, that tip velocity will add to the deceleration whereas if the tip is at state 3 and moving to the left, that tip velocity will subtract from the deceleration.

Real systems ramp the velocity change over some brief interval and I would advise you to include the ramp up and down times in the simulation because that will help the solver, which has a difficult time with an instantaneous change in velocity.

[fabrizio.pirozzi]

Hi, thanks a lot guys for your replies,

I attach an image to explain better the problem:

.

 

this is my assembly. It can rotate as shown in figure, thanks to the joint.

Now, imagine that the system initially is in horizontal position then moves to reach another position; when it reaches this final position I need to know how much time it needs to stop definitively itself . So, for now, I'm not interested about the oscillations during the transient, but only in the final position.

 

Thanks.

Fabrizio.

[peteroznewman]

Please provide more detail about which faces on the geometry have the rotation applied that moves the arm to a new position (angle). Does the cylindrical face have a bearing in it?

Also provide an angle vs time graph for the move.

[peteroznewman]

The oscillations during the move affect the oscillations during the stop. There is a complex relationship between where you stop and the oscillation response after you stop.  You can eliminate this complex relationship by having the transient analysis start from a steady rotational speed, and eliminate the ramp up from zero to full speed which causes the oscillations while moving that affect the oscillations after the stop.