CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Autodesk Simulation CFD (http://www.cfd-online.com/Forums/autodesk-simulation-cfd/)
-   -   heat transfer analysis (http://www.cfd-online.com/Forums/autodesk-simulation-cfd/95850-heat-transfer-analysis.html)

tolik January 5, 2012 05:29

heat transfer analysis
 
hello.

I am trying to perform an analysis on Counter flow HE because the high temperature differance (~150 K) the solver that I use is compressible, and the flow regime is laminar.
the BC are:
hot entrance: T=300 k;p=1000 psi; m=0.05 g/sec
hot exit: m=0.05 g/sec;
cold entrance: T=100 k; m=0.05 g/sec
cold exit: p=14.7 psi; m=0.05 g/sec.
working fluid : Nitrogen- user defined properties
looking for : stady state solution,hydrodynamic and thermal.

the solution process is divergies no matter how I refine the mesh and during the iterations befor the brack down the system fails to sustain the mass conservation law.

If enyone encounterd such problems? what can be done for stabilasing the solution?

what can be the posible problem?

thank all the repliers.

ryan.arnaudin January 6, 2012 16:42

Your setup is over constrained and/or you are not using a variable property material.

Check the help documents for proper inlet and outlet boundary condition pairing for your type of compressible flow simulation. For example, inlet mass flow rate may be specified with an 'unknown' condition at the outlet.

Also make sure you are not using a constant property material. This is accomplished differently depending what version you are using.

tolik January 8, 2012 02:31

Ryan hello

Thank you for your sugestion I will recheck the parameters again,. however the problem with 'unknown' boundry conditions is:
when I set this condition, the program failes to fullfill the mass conservation equation and I get differant mass flows at differant cross sections.
also it is obvious that the mass rate are an numerical error and not a part of the solution process because it diverges.

best regards

Fugacity January 11, 2012 10:24

Your problem is quite over-constrained. CFDesign assumes all boundaries are walls unless you tell it otherwise. If you specify an inlet mass flow and temperature, you probably only need a Pressure=0 outlet condition, otherwise the solver will diverge [once you specify an outlet surface, the program knows the flow will exit there, so you don't need to specify an outlet mass flow condition@ that surface]. Or you specify P_in and P_out and solve for mass flow.

What exactly are you trying to solve for - It looks like you have all the quantities you need already specified as boundary conditions; mass flows, pressure, and temperature are all specified so what do you expect CFDesign to tell you? The only unknown you've left is the cold exit temperature, which you should be able to figure out by hand...?

Look in the help files for guidance in specifying the boundary conditions and make sure you use Nitrogen with a variable Material Environment (see help files for guidance)

tolik January 11, 2012 12:21

thank you for the reply.

I am interested to find the pressure and temperature at hot exit, the pressure at cold entrance, temperature at the cold exit.
and the themperature profile.

at this moment the program meneges to solve the flow filed of the cold channle and to keep the mass flow rate at the spesified value(0.05 g/sec)

but at the hot channle it jump several magnitude( ~5 g/sec) and the velocity magnitude become not physical at the entrance of the channle after several iterations.

this happen after I changed the boundry condition to:
hot entrance: T=300 k;p=1000 psi; m=0.05 g/sec
hot exit: unknown instead of m=0.05 g/sec;
cold entrance: T=100 k; m=0.05 g/sec
cold exit: p=14.7 psi; m=0.05 g/sec.


All times are GMT -4. The time now is 01:27.