roryboreyalice
Mechanical
- Sep 9, 2015
- 1
Hello,
I am having an issue with excessive kinetic energy in my dynamic (non-thermal), linear elastic simulation which I have set-up in CAE.
It's basically a hard metal plate (CPS4R) given an initial velocity impacting a softer plastic plate (CPS4R) that's at rest with frictionless hard-contact. I am using COH2D4 to simulate a crack in the softer plastic plate at the opposite end from the impact. Back of the envelope calcs have a kinetic energy for the steel plate around 10 J. At the end of the simulation, if I look at ALLKE for the steel plate I see around 10 J at t=0 and it steadily drops off as the impact occurs over a span of 65e-6 (step 5e-9). The total energy of the COH2D4 set is less than 0.25 J, and is zero for the first 30e-6, everything seems good, acceleration and displacement maps look valid.
This is the odd part, while the simulation is running and I watch the monitor it shows a constant total energy value around 15 J, and an initial system KE of 15 J which hits a minimum of 10 J, this is confirmed with ETOTAL flat-lined at 15 for the entire simulation. However, if I sum ALLKE for all parts in my model for the entire simulation I do not exceed 10 J. More back of the envelope, if I consider that somehow the plastic plate *also* has a velocity equal to the steel plate at t=0 I only get a total energy of about 13.5 J.
If I extract all nodal velocities for all my parts, and sum the elemental kinetic energies I can generate plots that perfectly match the ALLKE data from Abaqus which makes perfect sense to me and is exactly what I expected.
Where is this extra kinetic energy coming from that abaqus is reporting in ETOTAL and in the monitor window under KE?
I am having an issue with excessive kinetic energy in my dynamic (non-thermal), linear elastic simulation which I have set-up in CAE.
It's basically a hard metal plate (CPS4R) given an initial velocity impacting a softer plastic plate (CPS4R) that's at rest with frictionless hard-contact. I am using COH2D4 to simulate a crack in the softer plastic plate at the opposite end from the impact. Back of the envelope calcs have a kinetic energy for the steel plate around 10 J. At the end of the simulation, if I look at ALLKE for the steel plate I see around 10 J at t=0 and it steadily drops off as the impact occurs over a span of 65e-6 (step 5e-9). The total energy of the COH2D4 set is less than 0.25 J, and is zero for the first 30e-6, everything seems good, acceleration and displacement maps look valid.
This is the odd part, while the simulation is running and I watch the monitor it shows a constant total energy value around 15 J, and an initial system KE of 15 J which hits a minimum of 10 J, this is confirmed with ETOTAL flat-lined at 15 for the entire simulation. However, if I sum ALLKE for all parts in my model for the entire simulation I do not exceed 10 J. More back of the envelope, if I consider that somehow the plastic plate *also* has a velocity equal to the steel plate at t=0 I only get a total energy of about 13.5 J.
If I extract all nodal velocities for all my parts, and sum the elemental kinetic energies I can generate plots that perfectly match the ALLKE data from Abaqus which makes perfect sense to me and is exactly what I expected.
Where is this extra kinetic energy coming from that abaqus is reporting in ETOTAL and in the monitor window under KE?