Spring element extension does not match theory.

I'm running a basic simulation of a 0.650-tonne mass point attached to a 5000 mm long linear spring element (PSPR) with a stiffness of $k =187.144$ N/mm. The gravitational acceleration is set to $g = 9807$ mm/s².

Using the formula $x = m*g / k$ , the expected static extension of the spring is approximately 34.073 mm. However, when running the simulation in Dytran, the spring stretches to 68.12 mm—almost double the theoretical value.

The same discrepancy occurs when replacing the spring with a ROD element, using $E = 1532$ MPa and $A = 61.07$ mm².

What could be causing this discrepancy?

Parents

0 Shank Kulkarni 3 months ago

Hi Gavril,

The reason for this discrepancy would be that the simulation is not being pure static. Dytran is an explicit solver, which means it is primarily designed for transient dynamic simulations (e.g., crash, impact, blast, drop tests). Explicit solvers do not perform true static equilibrium calculations as an implicit solver (like MSC Nastran SOL101 or Marc) would.

If your simulation is not purely static, the mass might be oscillating around the equilibrium position, especially if there's no damping. This can momentarily lead to larger displacements. Try adding damping or running the simulation for a longer time to see if it settles at the expected 34.073 mm.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel

Reply

0 Shank Kulkarni 3 months ago

Hi Gavril,

The reason for this discrepancy would be that the simulation is not being pure static. Dytran is an explicit solver, which means it is primarily designed for transient dynamic simulations (e.g., crash, impact, blast, drop tests). Explicit solvers do not perform true static equilibrium calculations as an implicit solver (like MSC Nastran SOL101 or Marc) would.

If your simulation is not purely static, the mass might be oscillating around the equilibrium position, especially if there's no damping. This can momentarily lead to larger displacements. Try adding damping or running the simulation for a longer time to see if it settles at the expected 34.073 mm.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel

Children

0 Gavril Birsanu 3 months ago in reply to Shank Kulkarni

Hi Shank! Thank you for your reply!

I should have mentioned the the system is damped and the value of 68.12 mm is obtained after everything settles.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Shank Kulkarni 3 months ago in reply to Gavril Birsanu

That's interesting. Then another possibility would be that the model is overdamped. It might seem like everything is settled down but may not have actually reached steady state. How are you applying the damping? And if you plot the deformation vs time graph, how does that compare with the theoretical one?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel