Learning Forum

[neilclyde]

Hello, my aim is to simulate the stress and deformations of an assembled styrofoam box at impact landing. (performing this analysis before using it for the Balloon launch payload) I applied all the necessary prerequisite boundary conditions required, the analysis does run but the results in the form of stresses and deformations just do not show. at all as intended Please helpnBasically, I do not even get to see the system falling down onto the floor. I applied velocity as initial conditions, fixed-support, a decent mesh, material allocation and suppressed contacts cuz I did not want the bonded conditions to prevailnn

[Ram Gopisetti]

.

Hi neilclyde , Based on what you have posted, i can suggest you following recommendations.

1) Increase the end time of your solution, 1e-8 is very less, try 1e-3 or in such order to see fair amount of deformation and relaxations in the model.

2) Since you have used Initial conditions as Velocity, I hope your impacting body is in contact with the bottom floor, Mostly you don't see much change in the movement of the box unless you have maintain certain distance between the bodies and this would increase the simulation time.

If you are new to the explicit simulation studies i recommend you to go through the AIC courses in the following like to know the formulations and best practices in simulating these kind of impacting problems.

Cheers, Ram

.

[neilclyde]

Hello there, Ram. Yes, I recently became acquainted with Explicit Dynamics. I'll look through the videos you attached and get back to you! Thank you very much for replying to my query! : D

[neilclyde]

Hello Array , I went through the videos and attempted the simulation again. I used PS foam as material for the box parts (flexible) and concrete for the flooring (rigid). The characteristic length in my case(h) = 26mm as I assumed it as the element mesh size. c = sqrt(Youngs's modulus/longitudinal wave propagation speed) = sqrt(1.72e10/1045) = 1245m/sec, f = 0.9. From these, I thus get the time-step size as 1.88x10^-5secs. Yes, in the design, the bottom portion of the box assembly does touch the floor. When I hit run, I get a message in 5 mins that the time step is too small. And again, nothing moves and the result values are too small to be valid : ( I closely took the help of the can-dye impact problem. Is it may be a problem with my Solidworks assembly, they're just 6 sides of the box assembled via mates.

[Ram Gopisetti]

Hi Array , its good to see your work on time step size but how about end time of your simulation is it same as time step ( i don’t think its one and same) or do you have any estimates on that please do specify.
Cheers, Ram

[neilclyde]

hello Array , I kept the end time size as 1e-003. Well, actually I changed it from the default 1e-008 to that on your recommendation. In my case, they are not the same! : ) Still, results are not what I expect to see. Can you please tell me how to decide the end time step size value? And thanks a lot! I deleted the default contacts and implemented the frictionless boundary set option active.

I am also trying the same simulation using the Solidworks Drop test, which also uses the Explicit time integration method, and the solution time after impact is around 880 microseconds which I obtained depending on the dimensions and material of my model. At the moment, the analysis is 10 hrs in, and at 35%? I know that my mesh is really fine, but I am giving it a shot to see how the results compare in the end, and if deformations show on the model.

[Ram Gopisetti]

Hi Array , you choose the end time to accommodate the event and the relaxation of the objects. Have you choose Mass scaling to speed up the simulation. Having multi body shows considerable effect on the solution time and guess what its normal and we have to give time to solve. Drop down your results after you finish it for further changes.
Cheers, Ram

[neilclyde]

hello Array , and thanks a ton for getting back. Ahh, mass scaling. Oh good to know, I'll figure it out, thanks! The Solidworks simulation was a success but took 1 day, 10hrs to solve. But I'm pretty happy with the results. Having gotten some success, I will carry out simulations in Ansys and then get back ?
Cheers!

[Ram Gopisetti]

Good on you, Array , Have you checked the material models' that you have used will accommodate the nonlinear behavior of the body(ies) on such short term loading. Such as having such a fall , did your chosen solver predicted any dents and unwanted bends or breakage? do consider such things which are must in impact analysis.
Cheers, Ram

[neilclyde]

.

hello ram_gopisetti, for the appropriate non-linear effects I applied the plastic von misses material model and no-penetration contacts to the falling object assembly parts (and not the linear elastic material model) that the large deformations such as dents and breakages get captured ?? From what I learned so far, linear elastic model obeys Hooke's law and deformations would thus be obeyed only until the proportional limit and nothing after. At the moment I am applying sensor points to help capture stresses at custom regions. But here's a snapshot from yesterday. Besides, I am figuring out ANSYS as well and shall revert when I need assistance?

.

[neilclyde]

.

hello ram_gopisetti , in Ansys I have been getting large energy balance errors and the solutions run very quickly, however, the results obtained are really incorrect. My mesh element quality is as shown below. Is it because I need to eliminate these mesh errors and thus bring mesh quality higher, only then the energy imbalance problem would disappear? What could be the reasons, kind sir? Also, my minimum characteristic length value is really really small as you see

.