Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Time steps in Explicit Dynamics Analysis Using Altair Radioss?

Hi ,
I am just started the Radioss solver using Radioss Elearning by Altair One.
I have little bit confusion about time stepping.
What is Critical Time Step, Element Time step & Nodal Time ?
∆𝑡 <∆𝑡c, Here what is ∆𝑡 and ∆𝑡c ?
I know ∆𝑡e is Element time step.

Thanks & Regards,
Sivaprakash V.

Find more posts tagged with

HyperMesh

Radioss

SimLab

Accepted answers

PaulAltair

∆𝑡 is the timestep at which you are running (or attempting to run at)

∆𝑡c is the 'critical timestep', this is the smallest timestep at which the model will run stably with no intervention, this is the smallest elemental/nodal timestep anywhere in the modelWe typically calculate the timestep manually using elemental timestep which is based on element size / speed of sound in material, Radioss uses nodal timestep internally which is based on nodal stiffness and nodal mass, but is the same as elemental timestep for a regular element.Speed of sound in a material is square root of modulus over density, so, e.g. for steel it is 5172194 mm/sThis gives the classic explicit example of stable timestep for a 5mm element in steel being approx 1e-6s (5/5172194 = 9.667e-7)In practice the smallest timestep in your model will often be attached to the smallest element, but since modulus and density play a part, a larger, but stiffer/lighter element may be the critical element. Basically, the smaller, lighter, stiffer an element is, the smaller its timestep.When you run Radioss starter (0000.rad) the 0000.out file gives a summary at the end of the timestep sizes in your model and an estimation of what levels of mass scaling would result from choosing a larger timestep.You may use /DT/NODA/CST in the engine (0001.rad) to run with a larger timestep at the expense of adding mass

All comments

PaulAltair

In the notation: delta t is timestep, delta tc is critical timestep. Radioss uses nodal timestep internally, which for an irregular element may be slightly different to the elemental timestep of the element it belongs to, but you may consider here to be broadly equivalent in most circumstances. The model critical timestep is the single smallest nodal/elemental timestep in the model (i.e. the timestep at which the model could run stably with no intervention of adding mass) it is related to sound speed and element size, sound speed is (square root of modulus over density) which will give a velocity, then dividing the minimum characteristic length of element by that velocity gives critical/stable timestep, the classic example is that a 5mm steel element has a timestep of approximately 1e-6s (square root of 210000/7.85e-9 being 5172194, so 5mm/5172194 = 9.667 e-7 s) that is in T/mm/s untis, but the calculation works for any consistent unit set, if you were using ms, it would be 1e-3 approx. Typically the smallest element in your model will attract the smallest timestep, but because modulus density plays a part, it could also be a larger element with a stiffer/lighter material. Essentially the smaller, stiffer, lighter and element, the smaller its timestep will be. Other model entities like contacts/springs etc may also influence timestep due to their stiffnesses.

PaulAltair

∆𝑡 is the timestep at which you are running (or attempting to run at)

Sivaprakash_V

Hi,

Thanks for your answers.
I will go through the suggestion what you mentioned in the above comments.

Thanks & Regards,
Sivaprakash V.