Learning Forum

[2281935340]

Hello everyone,

I am new to Ansys. And i am currently simulating laser additive manufacturing via Ansys Workbench (transient thermal coupled with static structural). In the simulation I need to use birth and death elements. In Ansys Workbench it is possible to control the birth and death of a element by the step. But i would like to know how to control the birth and death of the elements by the simulation results (temperature). It would be nice to have some examples.

Thanks in advance!

Ky

[dlooman]

This is possible in principle with APDL scripting, but might not be a good thing to try as a new user. If you want to try it anyway, make a very small cube model to test it on. The APDL outline is as follows:
/solu
*do,time_val,0,1,1e-3
!!! Apply load
solve
nsel,s,temp,,0,1000 ! select nodes below 1000 degC
esln ! select attached elements
ealive,all
allsel
*enddo
fini

[2281935340]

Hello dlooman There are a few more things I would like to ask you about your APDL command.
In the command "!!!! Apply load" means here I need to add the load (laser) as a command stream?
Don't I need to kill the birth and dead elements first when using your command stream to control them?
For example:
/solu
*do,time_val,0,1,1e-3
ekill, element1
ekill, element2
ekill, element3
(Elements 1 to 3 are the elements of birth and death that I have named myself by "Create Named Selection".)
!!! Apply load (laser)
solve
nsel,s,temp,,0,1000 ! select nodes below 1000 degC
esln ! select attached elements
ealive,all
allsel
*enddo
fini

Thanks in advance!
Ky

[2281935340]

I added your APDL command to try but it failed. I was hoping the elements would be dead at first. When the elements are heated to a certain temperature by the laser they come back to life. I tried to modify on your command also tried to write the command like the information on the internet. But unfortunately both attempts failed. In the simulation, the previously set dead elements stayed dead. I verified the command of my laser. The command for the laser was correct. I would like to ask you if you have any other ideas on this issue?
Modifications based on your command:
ekill,s1
ekill,s2
ekill,s3
ekill,s4
ekill,s5
ekill,s6
ekill,s7
ekill,s8
ekill,s9
ekill,s10
/solu
!!!command for laser
solve
nsel,s,temp,,500,10000 ! select nodes higher than 500 degC
esln ! select attached elements
ealive,all
allsel
*enddo
fini
Modifications based on the information from Internet:
ekill,s1
ekill,s2
ekill,s3
ekill,s4
ekill,s5
ekill,s6
ekill,s7
ekill,s8
ekill,s9
ekill,s10
/solu
!!!command for laser
solve
finish
!
/POST1
set,...
etable, temper,temp
esel,s,etab,temper,500
finish
!
/solu
antype,,rest
ealive,all
esel,all
*enddo
fini

Thanks in advance!
Ky

[Ollieturpin]

I have a similar concern in that I am trying have elements deactivated when they exceed a certain strain value, currently my problem is that the code is telling me that when I try to request the stress or strain data to determine which elements are within the range that need to be killed that the data is unavailable. I have attached the code that I am using. Some elements of it are slightly off as I have commented some stuff out resulting in excess endif's.

/com, #############################################
/com, ##£
!*get,myncmls,active,,solu,ncmls ! cumulative number of load steps
!*if,myncmls,gt,0,then ! if this is after the first
cmsel,s,Cancellous ! ELEMENT component to be checked
nsle ! nodes on these elements
SELTOL,1.0e-12 ! stay within range below
! #########
nsel,R,S,eqv,1E8,1E11 ! re-select nodes with results above a value <<<
! #########
!SELTOL ! selection tolerance back to default
*get,numnode,node,,count ! how many nodes with result above?
*if,numnode,gt,0,then
ESLN,R,1,ALL ! elements with any node selected
*get,numelem,elem,,count
*if,numelem,gt,0,then
NROPT,FULL
ekill,all
SHPP,ON
*endif
*endif
allsel
*else
! if first load step, have result written to in-memory database
fini
/config,noeldb,0 ! write results into the database
/solu ! continue with solution
*endif
/com, ##£
/com, #############################################

If anyone has any advice to solve my problem that would be a lot of help.