CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   CFX Casting problem: Can it be done all in one? (http://www.cfd-online.com/Forums/cfx/80042-cfx-casting-problem-can-done-all-one.html)

jfasl September 13, 2010 17:17

CFX Casting problem: Can it be done all in one?
 
Hi,

I have been using FLOW-3D for roughly 3 months now to analyze casting problems with great success, however, a new problem I have undertaken goes beyond the scope of what FLOW-3D is structurally capable of solving and I was hoping CFX/thermal/structural could solve at least part if not all of the problem. It is simply this:

Step 1) Pouring liquid metal from a crucible into a ceramic basin shaped container and letting it solidify (initially the fluid could also be just "sitting" in the basin at the high temperature if it makes the simulation easier)

Step 2) After the liquid-solid change, the metal then goes through 2 solid-solid phase transitions

Preliminary experimental findings show cracks forming in the ceramic due to stresses caused by expansion of the metal due to solidification (similar to expansion upon changing water to ice). I hope to be able to show the stresses in the basin due to the casting. I had a couple of different ideas in order be able to do this, please comment if they seem absurd.

For Step One: Do this in FLOW-3D, get a temperature distribution and manually enter it into ANSYS sturctural once the metal is solidified (this seems like it would have a lot of error associated with taking results from one program to another though hence why I wondered if I could do this step in CFX and input it as a body condition in ANSYS thermal). If I can do it in CFX, I thought about having temperature depedant viscosity that basically goes infinite at the solidus temperature of the molten metal, but I couldn't figure out where to input temperature dependant fluid properties in CFX.

For Step Two: Have temperature dependant Cp and thermal expansion values that "spike" at solid-solid phase transitions, simulating release of latent heat, although I haven't figured out hwo to do this in ANSYS yet, help would be appreciated.

Any thoughts or comments to help me out would be appreicated. This seems like a much more difficult problem than anything I've done thus far, and I haven't used ANSYS for anything other than strutural and thermal analyses up to this point, thank you for your time.

ghorrocks September 14, 2010 03:14

I think you are using the wrong software. You need something like Thercast from Transvalor - http://www.transvalor.com/thercast_gb.php I am pretty sure it can do everything you are looking for. I think you may also be able to model the phase transitions during cool down also - but not sure on that, ask them. I used their Forge3 software in a previous life and found it excellent for its intended purpose (ie forging)


All times are GMT -4. The time now is 21:58.