101 of casting simulations 1

[bjpil]

Hello all,

I am hoping this is the best place to ask for help. I’ve been using FEA for a while and now I’ve been asked to critique a casting simulation using a specific flow analysis software. How much different is structural FEA from flow analysis, besides the obvious, and how does one judge the quality of a casting simulation model. As a starting point:

1) Should I be checking for convergence?
2) how important is mesh quality?
3) does velocity give me any real information? (Q=VA)
4) does the pressure output tell me anything? (delta P)
5) how should I correlation the results?
6) are boundary conditions as critical as they are for structural?
7) what should I be doing to make sure the results are the best that can be predicted?
8) or is flow analysis just used to see how well the cavity fills?

The tool is presently used for just #8 with no regards to the other 7 but I believe that it could be used for more. Any suggestions would be appreciated.

Thanks,

BJP

 

[GregLocock]

I'm no casting expert but the pressure and velocity plots are telling you where the material will have difficulty flowing, and where porosity is likely to initiate.

As to the rest I rather think that casting analysis is an order of magnitude more difficult than structural linear FEA, if only because of the difficulty of getting correlation data.



Cheers

Greg Locock

 

[mloew]

BJP,

I am also not an expert, but have general awareness of the commercial tools for casting simulations. Without knowing the specifics of the tools, I would state the following:
[li]In any simulation, convergence is important. How could it not be?[/li][li]Correlation might be handled by comparing intermediate simulations to incomplete (interrupted) castings.[/li][li]Boundary conditions are always very important.[/li]

You may also want to look at the thread Aluminum Casting Simulation Software (thread330-66671) where some links to commercial codes were listed.


Good luck!

Best regards,

Matthew Ian Loew
"Luck is the residue of design."
Branch Rickey

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[YYS]

Some quick comments:

1) Should I be checking for convergence?

For numerical solutions, only the converged solutions make sense. Most commercial codes lets you choose a convergence criterion and level. You can check the convergence with these tools.

2) how important is mesh quality?

Like bricks to a house.

3) does velocity give me any real information? (Q=VA)

Metal flow determines the temperature and mass distribution in a cavity. It also helps determine if there is too much turbulence and possibly inclusions are traped.

4) does the pressure output tell me anything? (delta P)

"delta P" in turn determines the metal flow. A smooth filling requires smooth pressure change.

5) how should I correlation the results?

Flow - temperature - solidification rate - possible shrinkage? etc.

6) are boundary conditions as critical as they are for structural?

As critical and perhaps harder to determine.

7) what should I be doing to make sure the results are the best that can be predicted?

For thermal field, implant some thermal couples in die to measure and compare temperature with the calculated results.

For flow, use the same thermal couples and have their tips exposed into cavity to record the metal front arrival, then compare with the model.

8) or is flow analysis just used to see how well the cavity fills?

and if there is non fills etc.

 

[bjpil]

YYS,

Thanks for your reply. It was informative. I guess the best thing for me to do is run a few iterations of a simple casting simulation and see how the different models behave.

Checking with our flow software provider they said we should not have to worry about convergence because their software automatically adjust for it. They also suggest only two elemnts across in thickness is required, less than 20 bad elements (jacobian) in the entire model is a good model.

The reason I asked about velovity and pressure was to determine which one was more helpful for runner and overflow design optimization.

Another technique for correlation is to do short shots or partial shots.

BJP