DFLUX in transient thermal calculations

1

Hi all,

When DFLUX is applied to one C3D8 element in transient thermal calculations, strange things happen…

I provide the example file with just one element:

*NODE, NSET=NALL
1,0.0,-1.0,0.0
2,1.0,-1.0,0.0
3,1.0,0.0,0.0
4,0.0,0.0,0.0
5,0.0,-1.0,1.0
6,1.0,-1.0,1.0
7,1.0,0.0,1.0
8,0.0,0.0,1.0
*ELEMENT, TYPE=C3D8, ELSET=EALL
1,1,2,3,4,5,6,7,8
*SOLID SECTION, ELSET=EALL, MATERIAL=MID1
*MATERIAL,NAME=MID1
*CONDUCTIVITY, TYPE=ISO
1.0
*SPECIFIC HEAT
1.0
*DENSITY
1.0
**
*INITIAL CONDITIONS, TYPE=TEMPERATURE
NALL, 0.0
*STEP, INC=100000
*HEAT TRANSFER, SOLVER=ITERATIVE SCALING, DIRECT
1.e-5,1.e-3
*DFLUX
1,S1,100.0
*NODE PRINT,NSET=NALL
NT,RFL
*END STEP

Here the bottom face of the element is heated and the top face is heat insulated.

The results are following:

temperatures for set NALL and time 0.1000000E-04

     1  3.999640E-03
     2  3.999640E-03
     3  3.999640E-03
     4  3.999640E-03
     5 -1.999640E-03
     6 -1.999640E-03
     7 -1.999640E-03
     8 -1.999640E-03

heat generation for set NALL and time 0.1000000E-04

     1  1.499820E-03
     2  1.499820E-03
     3  1.499820E-03
     4  1.499820E-03
     5 -1.499820E-03
     6 -1.499820E-03
     7 -1.499820E-03
     8 -1.499820E-03

…

I didn’t expect negative temperatures at the top face. Also, heat generation values values at the insulated top face are negative. The magnitudes themselves don’t agree with the flux specified in DFLUX card.

For reference, I’ve run verification test oneel20df and the results are correct. I’ve played with different options, types and shapes of the element and apparently the cause is transient calculation.

Am I doing something wrong? Or maybe I misunderstood semantics of the DFLUX card.

WBR, Vadim

2

I’ve replaced DFLUX card with

*CFLUX
1,11,25.0
2,11,25.0
3,11,25.0
4,11,25.0

And results are the same. So, it’s not about DFLUX, but I rather don’t understand transient heat solver quite well.

3

Hi Vadim,

I looked into your problem. The cause is indeed the transient aspect of the calculation. Due to the heat input a kind of thermal front is going from one side to the other. With just one element it is not possible to model this thermal front. Using a lot more elements should solve the problem.

Best Greetings,

Guido

1 Like