Learning Forum

[Max Power]

Hi there!

I was wondering 2 things:

a) Is it somehow possible to get the time of a transient solution during a solution? Similar as you can get a the number of substeps with *GET,something,ACTIVE,,SOLU,NCMSS (or NCMLS)

b) Is there a way to execute an *IF statement at the beginning of every substep?

I know it might be computationally bothersome, but it is more out of interest.

Reason behind this is that I have logical information stored in an array and I would like for it to be checked against the time of the transient solution without exiting the /SOLU.

 

Thanks for any input! 

Regards,

Max

[jj77]

You should be able to use that *get as it seems to be solution related.

The *if statement might go in a do loop in a transient analysis.

(I can not be 100% since I never used this,but try it)

 

On how to do this (*do loop) see this chapter in help:

5.6. Solving Multiple Load Steps

[Max Power]

Hi jj77

Thanks for you input!

My question regarding the *GET command was, if anyone knows if there is an *GET command to get the solution time. In the help no solution option for the *GET command in relation to the solution time is mentioned. The example I gave is for the number of substeps. Something along the lines of *GET,someName,ACTIVE,,SET,TIME - but this is listed as an postprocessing option. I don't know if it will work during the solution.

Thanks for your suggestion with the *DO loop. I was trying to avoid a fixed time step solution, because of the added computational cost to my simulation, but if I don't find any other way this would be a solution.

 

[jj77]

In the do loop you control the tm_increment and all time parameters so you know what time it is when the next iteration in the loop starts.

 

Now I have never used this method so you will need to try things out.

 

[Max Power]

Interesting thought...If possible i wanted to keep the automatic time stepping on, but maybe I can at least match it to the previous calculation.

Reason behind my question is that I have a one way coupled thermal-mechanical simulation and I want to execute an EKILL command with the information from the thermal step, possibly without exciting the solution part of the mechanical step. As of now I only saw a possibilty to do it with a fixed-time step calculation or with a rather big number of load steps, with both not really being ideal.