How to input the time period to get the results of 1-year.. etc

[Chalmers2009]

Hello all,

I want to run my model through coupled temperature-displacement analysis to check the results of 1-year, 2-years and so on. Can anyone guide me how I may set the time increment in the step to get the said results.

Thanks in advance.

 

[DrBwts]

WOW! That's a long time to run a simulation. Here's my best guess, calculate a consistent set of units so that 1 unit of time as Abaqus sees it is 1 year.

I'm no Civil Engineer though, there may well be a way to do this, maybe some implicit method?

 

[Mustaine3]

I'm not sure what you're doing, but be aware that Abaqus is no fatigue code. It has limited capabilities regarding low-cycle fatigue, but that's it.

 

[Chalmers2009]

@DrBwts, thanks for your response. Actually in the Step window, I have to define Time Period and Incrementation sizes. So far I understand that I can give No. of seconds of 1-year in time period and make a suitable incremental order for it. But I am not sure whether it is right or wrong.
@Mustaine3, Thanks for your comment. Actually I am running a thermo-mechanical coupled model and wanted to examine the results of 1-year... More clearly, a heater rod is inserted at some depth in the rock. The heater maintains it temperature at 100°C for the whole analysis period. As soon as time passes, the surrounding rocks of the heater will be heated up and the stresses in that location will be changed due to expansion of the rock. I need to calculate the stresses and temperature at designated point around the heater for 1-year, 2-years and onward..

 

[DrBwts]

How are you modelling it? Using the implicit or explicit solver?

If you are using the explicit solver then it would be quicker to run the actual experiment than the computer model.

 

[Chalmers2009]

@DrBwts, Thanks for your comment. If you have any other suggestion regarding time period, please share with me.

 

[IceBreakerSours]

My two cents:

Clearly, running the time domain problem in 'physical' time is not practical. In all likelihood, you'd want to run the problem in 'simulated' time OR the frequency domain. Either way, damage as a function of 'simulated' time or frequency must show up in the material law.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[corus]

I doubt that you'd need to run the analysis for a 1 year time period as it'd have reached steady state long before that, and at a guess in a matter of days. In addition you'll probably find that the maximum temperature change, and hence maximum stresses will occur in the first initial period. If you use periods of 1 year or so then you may find the initial time step and increments may miss these peak stresses. I'd run a simple analysis to gauge the change in stresses and temperatures before deciding on yearly periods.

 

[Chalmers2009]

@IceBreakerSours, thanks for your input. Actually I want to verify an experimental results through my simulation. The process wouldn't be end with any kind of failure. The main variables here are only the temperature and stresses.

 

[IceBreakerSours]

What makes you think you should run the analysis for years worth of physical time? I agree with corus (and others).

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[Chalmers2009]

Dear @Corus and @IceBreakerSours, Thank you very much for your assistance. I am trying to follow the guidelines you suggested.