TOTAL ENERGY PLOT

Hi Harshit,

Check what is the contributing more to the total energy. Check whether the contact energy is within the limit or high also check the hour glass energy.

If the problem is with hour glass energy, change the element formulation to fully integrated. 

If the problem is with contact energy, try changing the contact parameters.

Contact Energy is very less.

<?xml version="1.0" encoding="UTF-8"?>

Hourglass is under 10%limit of total

<?xml version="1.0" encoding="UTF-8"?>

here is complete plot of graphs<?xml version="1.0" encoding="UTF-8"?>

as you can see the internal energy absorbed by the structure is way more than total/ kinetic energy when impact happened, the impact happened at .02 sec from where kinetic energy starts declining. 

the friction co-efficient is .2 its a planar rwall. 

Somewhere from graph i infer that the kinetic energy of body is not decreasing as the slope of total energy almost remains constant before and after impact.

Models with initial KE (initial velocity) will have large Total Energy value but it should be somewhat constant throughout. But when the Total Energy varies significantly there is some major error in the model created or defined.

yes , even i think so, but i failed to identify as such , moreover i have recreated BCs many time but cant get rid of it.  what can be preferred checks for same?

Can you upload the model file?

yes,

i'm attaching the.rad file

Unable to find an attachment - read this blog

any observation Sir?

Try running it with the boundary condition INITIAL VELOCITY (/INIVEL). You have imposed /INIVEL/AXIS.

Try with initial velocity and see the output result

the body is freely falling after losing stability from a height from a tilted platform to ground 800 mm below the tilting axis in height

I have calculated angular velocity just above ground using parallel axis theorem and imposed same.

So imposed velocity may not apply real scenario.

Dear Harshitpuri,

What about the gravity you have applied to the model?

Hi Harshit,

For angular velocity calculations, what mass have you considered?

is it Kerb/curb mass of the vehicle, if yes, is it same as the mass of the FE model?

Since I am using 12.0 i am not able to open your model file, what is the material model have you used?

Any contacts, 1D elements?

Hello Prakash, 

Yes, i have taken the kerb weight for the calculation of angular velocity i have taken Moment of Inertia (xx) axis of rotation about  abt c.g 

and then used formula .5*(I+mr^2)= mh(hdiff)

r is distance of cg from rotation axis 

hdiff  is height diffrence (pre and post rotation to ground from point of unstability)

m is kerb weight of vehicle

there's no other mass in whole model even the tilting axis weighs less than a k.g (the mass of Fe model and bus are same 8.3 ton)

the angular velocity is around 2rad/s which is justified as i have carried out simulations on these values for the bus rollover analysis earlier.

contacts - all nodes are slave to rigid wall and have tried simulation with and without other contacts.

No 1 d elements in model

rigid and admass elements are present and they are assigned with boundary Condtion.

Dear Harshitpuri,

 

What about the gravity you have applied to the model?

Hello Ram,

Yes gravity is applied to the whole bus using curve except for stopper /snug

x 0

y -9810

x 1

y -9810

using grav load card

direction 

Z is direction of height (+ve Upwards)

Hi Harshit,

Can you export the model to RADIOSS 110 format and share the same to my dropbox link (in my signature below)?

Hi Harshit,

Can you upload the model in format as prakash mentioned above?

yes ram here it is 

prakash i have uploaded the same in your dropbox

Unable to find an attachment - read this blog

Dear Harshit,

1.In your model there are free nodes,

2. In rigid elements , Rbody ID is different from master ID

3. what is that component in the name 1?? It contains admas (mass) but you have not updated it as Rbody/admas in the component collector.

4.Check the model for intersection & penetration.

I can see few modelling errors. There are penetrations. Please correct them.

Try to plot energy balance curve for individual components to check which part is contributing to the undesired behaviour.

Please can you explain, how to plot energy balance curve for individual components..?

Hi Shriram,

From Analysis> select output block and choose comps from the selection menu and select all the parts, and thereby you can plot the time history for parts.

Please view this video tutorial at https://www.youtube.com/watch?v=Ef7Spn-j7fA&index=6&list=PL1u26y75SCrA_BOCLaSewVCNY7xRvCTnA where all these steps are shown.