[ghorrocks]

These functions simply define a region inside a sphere at an XYZ location. You would use this to define an initial condition of a spherical drop of stuff. And as you say, if you are using it as an initial condition for a free surface model it is good to put a smear function over it to smooth it out.

[Danial Q]

What is this delta(smearing distance) thing, u define delta on what basis? or it is just some random value?

[ghorrocks]

You have to write it as a CEL expression. They can be tricky to write. I think the documentation has some information on this.

[Danial Q]

yeah it is CEL EXPRESSION, in documentation as initial condition but documents does not tell any thing about, how DELTA is set just assumes some value from no where.

[Danial Q]

How can I animate the impacting droplet particularly? I have run animations for different variables study but still unable to visualize the impact on surface. Any suggestions please?

[ghorrocks]

Once the simulation is done this is a basic post processing operation. Have a look at the CFD-Post tutorials, and many of the CFX tutorials also include this sort of thing.

[Danial Q]

I am still unable to do phase change, what could be posssible reasons while functions are working fine. Temp, cp, density etc are getting reasonable values too.

[Danial Q]

I have set these pressures for domain and opening boundaries;
For domain: static P> relative P = 0 Pa
For opening : Relative pressure = 0 Pa
And Rerence Pressure = 1 atm
but when I post processed the reults, OPENING min and maximum Pressure shown are as follows;
Min P at opening = -16326Pa, Max = 0 Pa [global and hybrid/conserv]
Min P at opening = -75 Pa , Max = -74 Pa [Local and conservative]
Why is that?? while system is working at ambient pressure 1 atm. It looks like,its working in some space out of thsi world .

[ghorrocks]

Your post #67 is a question on accuracy - see the FAQ http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

I do not understand your question in post #68.

[Danial Q]

1 > I have checked domain imbalances and they are all in range of (0.001 to 0.05) % ,ofcourse less than 1%. So, there is no doubt that solution is converged successfully. am I right???
2 > Should I use finer mesh rather coarse one to improve the solution now??
3> I know that by applying symmetry boundaries, we model only half but if we have to model quarter domain, then what will be right steps?

Thanks

[Danial Q]

BACK TO SQUARE ONE!!! . I tried to contact some ANSYS support while providing them with details of model and got this reply;

" The setup described in your e-mail does not account for solidification in any way. Where is the heat of fusion included? What are you doing to bring the fluid to rest when it freezes? your formulation does not include solidification in any way. "

Q # How should I include Latent Heat of Fusion" in my model while properties are also defined as function of Temp???"
Pleaseeeeee!!!! Any suggestions? relevant Literature? Example or Tutorial????
Thanks

[ghorrocks]

I have never modelled solidifcation in CFX but I know people have done it. I suspect it is complex. You would have to ask CFX support for an example.

[ghorrocks]

As I said in the other thread solidification modellign is not a supported model in CFX. I know people have done it but I suspect it is quite complex to set up. I would not worry about any of your current convergence problems as you need to work out how you are going to model the solidification first.

[Danial Q]

My idea to solidify the liquid was, after hitting the substrate , liquid should cool down because of the temperature difference between fluid (2700K) and solid(300K) domains. If domains are automatically coupled through interface, heat transfer should occur and liquid should solidify.
Can we employ this idea if somehow, we could use the latent heat of liquid as heat souce for solid domain. Will it work??

[ghorrocks]

Of course, what you describe is the physics of solidification. But how are you going to implement it in CFX, given the tools it provides, and keep it accurate and numerically stable? It is the implementation, not the understanding of the basic physics which is the challenging bit.

[Danial Q]

I got following information regarding solidification modeling from some guy and I doubt it if its right approach,
1- define binary mixture for liquid solid nickle
2- apply thermal phase change model given in cfx
3- define user function for cp(T) and add latent heat in solid cp(for solidification) to make it liquid

does it make sense to u? I mean how latent heat can be added in cp, how thermal phase change model can be applied while air is there also?

I am asking you because I want to confirm the approach otherwise i will be wasting more time for nothing.

Thanks

[ghorrocks]

I have never modelled solidifcation so cannot say for sure.

But my understanding of binary mixtures is that they are for gas/liquid pairs, not liquid/solid pairs. Additionally CFX has no model for making liquid which solidifies not move any further. I have heard of using moving mesh to move the solidified material into a solid domain, and others keep it as a liquid but give it a very high viscosity to make it effectively frozen.

[Danial Q]

Yesss i found the way to do it.

[Danial Q]

Could you please tell me that how many conditions I can apply in a single nested if statement???? I tried this expressiom but there is error like (At least one operand of arithmetic operator is logical valued).

Error detected for '<=' operator at position 51.).

if(T>1726.5[K], if(T < 1725.5[K], if(1725.5[K]<=T<=1726.5[K], 620 [J kg^-1 K^-1], 595[J kg^-1 K^-1]), 145607.5[J kg^-1 K^-1]), 605.5[J kg^-1 K^-1])

what could be wrong with this, if i can use that many conditions in single expression? is this expression true for {T>1726.5 = 620Units}

[Lance]

You can use several nested if statements in one expression, dont remember exactly how many but way more than three anyway.

I dont think you need that third if statement, since it will already be
if T<1726.5 and T > 1725.5[K]

and I dont think CEL accepts expressions on the format 1725.5[K]<=T<=1726.5[K]