I am trying to solve the problem with various air properties(desity, diffusion coeff.) according to the time variation. Can someone please help me how to do? Thanks

Rhie

ps: I have phoenics 2.2

Dear SC Rhie

PHOENICS have implemented several models to handle fluid properties. If you are completlely sure your desired model is not included then continue to read this, otherwise stop and read POLIS entry for Fluid Properties.

It is simple but difficult to explain without to know how familiar you are with GROUND programming. Nevetheless we will try. OK?

I will assume you have a recomplilable PHOENICS version and you have designed your Q1 to solve a time dependent problem.

If you want to include a model for density (for example) insert this line into Q1 file

RHO1=GRND

Now you will have to assign your time dependency into GROUND. You will need the index ISTEP for the time step and the value of DT for time step size. You should add the proper code lines for store the total physical (or dimensionless) time. Moves into the Group 9 in GROUND (Section GRND, not GRND1, nor any GRND#) and insert the desired model. You will also need the index DEN1 for density.

Suppose I have a 2D problem with a cartesian grid and I want to increase the density every time step multiplying its value by a function "GFUNCT" of the elapsed time. I have the physical (total) time stored into TOTIME. Every time step I will update TOTIME and the value of function "GFUNCT"

TOTIME = TOTIME + DT GFUNCT = GFUNCT(TOTIME)

Then into the Group 9 (GRND density section) I will insert the following lines

L0DEN1=L0F(DEN1) DO IIX=1,NX

DO IIY=1,NY

INCELL=IIY+(IIX-1)*NY

F(L0DEN1+INCELL)=L0F(L0DEN1+INCELL)*GFUNCT

ENDDO ENDDO

Depending on how you have compiled your sources (I mean copiler options for variables) or how you declare the integer variable L0DEN1 you can move the line

L0DEN1=L0F(DEN1)

to Group 1 to save some time. If your problem is 3D the procedure is a little more complex but it is still easy to implement.

I hope this helps you. Regards

Kike

Is the density really changing in time alone, and not in response to other changes in the flow? Maybe so, but this strikes me as unusual. If this really is true, you need to do it in Ground, as Kike suggests.