Population of "file.rth" ?

[harshn]

Hi All,

Can anyone tell when file.rth gets populated during transient thermal analysis as I am need to pick up temperature distribution in the body at the end of transient analysis using ldread command.

Thanks and Regards

 

[Stringmaker]

You should be able to issue LDREAD anytime after SOLVE. Completion of the solve command indicates that load step has finished solving.

 

[harshn]

Hi,

@Stringmaker I tried the following set of commands, where I read a temperature of 500 K from the table. I am doing a transient thermal analysis in this complicated way as I need to stick in few other analysis after every cycle of transient analysis.

/PREP7
length=1.0
height=1.0
blc4,0,0,length, height

ET,1, PLANE55
MP,Dens,1,920
mp,c,1,2.040
mp,kxx,1,5
ESIZE,0.05
AMESH,ALL
FINISH

/SOLU
ANTYPE,STATIC

NSEL,S,LOC,Y,0
D,ALL,TEMP,100

NSEL,S,LOC,Y,height
D,ALL,TEMP,100

ALLSEL
SOLVE
FINISH

*Do,i,1,1,1
/SOLU
ANTYPE,TRANS
LDREAD,TEMP,LAST,,,2,,rth

time,300
nropt,full
lumpm,0
nsubst,10
neqit,100
autots,off
lnsrch,on
outres,all,all
kbc,0

temperature = 500
NSEL,S,LOC,Y,height
D,ALL,TEMP,temperature

NSEL,S,LOC,Y,0
D,ALL,TEMP,100

ALLSEL
Solve
Finish

*IF,i,Eq,4,Then
*Exit
*Endif
*Enddo


The confusion I am facing is in LDREAD command which reads the file.rth and apply the temperature as initial condition (use of option 2). Then I specify Boundary condition in transient analysis and use ramped loading then at the bottom of square Y=0 I expected that the temperature at these nodes will remain 100 K but when I plot the temperature variation on these nodes it ramps from zero to 100K rather than staying at 100K. Simillarly for top nodes the temperature ramps to 500 K from zero and not from initial condition which was applied using LDREAD.

Can some one tell me if I am using some option incorrectly?

Regards

 

[Stringmaker]

Harsh,
The LDREAD command is only valid for application of temperatures as body forces in a structural analysis. If you wish to define an initial temperature condition in a transient analysis I would look at the IC command. I'm not certain that there is a direct way to read an *.rth file in as initial conditions. In this case applying uniform temperature as an initial condition is simple. However, if you have a varying temperature distribution throughout you will need to use array parameters retrieve and apply those temps as initial conditions. One thing you will need to be aware of is that if you apply a temperature as an initial condition, applying a boundary conditon subsequently may overwrite the initial condition. Look at the DCUM command if this is the case. I hope this info helps.

-Brian

 

[harshn]

Hi Brian

The description of ldread command is as follows:
LDREAD, Lab, LSTEP, SBSTEP, TIME, KIMG, Fname, Ext,
Lab = TEMP --
Temperatures from a thermal analysis applied as body force loads [BF] in a structural analysis, an explicit dynamic analysis, or other type of analysis. Can also be temperatures applied as nodal loads or initial conditions (see KIMG description).

---> my interpretation for "other type of analysis" is any analysis ... and that is where the KIMG option helps in applying temperature in different ways:

When used with the TEMP label, KIMG indicates how temperatures are to be applied. In an explicit dynamic analysis, KIMG = 0 is the only valid option for applying temperature loads.

0 -- Apply temperatures as body loads
1 -- Apply temperatures as nodal loads
2 -- Apply temperatures as initial conditions

At points where I did not applied the boundary condition the transient analysis started from the initial condition which was imposed by ldread command.

I will search for more in DCUM command.

Regards
Harsh

 

[Stringmaker]

Harsh,
It seems as though the LDREAD command should work. Thanks for pointing out the KIMG attribute I overlooked that originally. When using the code you supplied I noticed that the LDREAD does not seem to have an effect as you mentioned. If you look at the output window after the command is issued it says the that temperature for zero of the selected nodes has been stored. I don't understand why this is. I don't see the LDREAD command mentioned as a suitable means for applying loads under the transient thermal documentation either. The IC command is recommended there. Have you given this approach a try?

-Brian