basic questions about time in ABQ

[alexzive]

hello,
my question is:
1) if I choose second as time unit (e.g. in heat transfer analysis --> flux in J/s/mm²) then my single time step in the simulation corresponds to one second, right?
(ps: I am using ABQ-STANDARD)
2) if I take 2 simulations both referring to the same model, both performed in ABQ-STD, both covering the same "time period" (e.g. 1000 time steps=1000 sec?) but with different final total number of increments (for numerical reasons) and I want to compare the results at the same "time" (e.g. temperature fields across the region), is it correct to take 2 snapshots at the same "step time", e.g. step time = 57?

thank you
Alex

 

[xerf]

Some remarks:
1. Heat transfer : transient or steady state ? In steady-state heat transfer the time is used only to prescribe the loading path.
2. If you use automatic time incrementation you might not get "snapshots" at the same time. It is better to compare the results at the end of the step.
3. Your post is a bit confusing: did you create 1000 steps ? Or you ar referring to 1000 fixed time increments during 1 step ?

 

[alexzive]

sorry to reply you so late and thank you for your reply. My questions are still open.

to your questions:

1) Transient
2) ok, you are right.
3) ok, I try to explain it better.

I am looking at the transient and try to make a comparison between different heat tranfer TRANSIENT simulations on the same geom. model, varying for example the heat flux . So I am interested to comapre "temperature field snapshots", but I have to be sure I am comparing at the same (REAL) time in second. I need SECONDS and not TIME STEP as time units.
How to be sure that I am actually working with seconds and not with milliseconds? I know that you decide the unit and they should remain coherent within a simulation. Is everything related to the value you give for the Thermal Conductivity [JT–1L–1 –1] or the Physical Constants (like Stephan Boltzmann constant - I am simulating Heat Radiation too).
To my understanding, If I put such constants in seconds, everything will be in seconds, for example time step n. 50 lasted 0.4 s right?

Many Thanks , Alex

 

[alexzive]

errata for my previous post+reply:

with time-step I mean time INCREMENT within a single STEP (sorry but also in Abaqus-Visualization when you plot an animation it shows "step time" in the state block meaning time increment)

 

[mrgoldthorpe]

alexzive,
As you said in your original post and above, when you do a heat transfer analysis you set the ABAQUS time unit to seconds by your choice of physical units. So the *total* ABAQUS time will be the *total* real time in seconds in your case. (If you were doing a creep stress analysis, on the other hand, you would probably find it more convenient to put the creep law units in terms of hours. because that's the convention. In that case ABAQUS total time would correspond to total real time in hours.)

STEP TIME is the accumulated ABAQUS time in the current step you are viewing. That means it's the total accumulated real time *within the current step*. It seems as if you are using a single step to carry out both thermal transient analyses. If that's the case, you can make the comparisons you want at the same STEP TIME since that would be the same real time, regardless of the increment number. As xerf points out, using automatic time incrementation means the two analyses may never quite coincide at the end of any pair of increments, so it's better to use more than one step and make comparisons at the end of each step. At the final increment of any step, the total accumulated *real* time in seconds is the total accumulated ABAQUS time for the steps completed so far.

Sorry to labour this, but I hope it clears it up for you.

 

[alexzive]

everything is now clear Mr. Goldthorpe,
thank you very much, Alex