Excel CFD 2

[Enginerd9]

I already know what you're thinking... why? Maybe even "how?"

Independant work, using the tools I have available on my home laptop. I'm not wealthy enough for the fancy software packages for personal use.

I have found a nice solver algorithm package as a good starting point, but I want more. I understand and have written my own Navier-Stokes equations / solver in MatLab before, but no longer have access to that software. Do any of you know some macros or VBA code for running simulations in Excel? Or even a really good refresher publication for how to set up the equations and get them ready for programming?

Conditions would be simple: steady state, incompressible flow at low reynolds numbers. The idea is to calculate the drag on a submerged object in water, for example.

Thanks!

 

[cowski]

Maybe you can upgrade the tools on your laptop inexpensively; have you had a look at CAElinux? It has modules for meshing, FEA, and CFD. It is open source in case you want to inspect the inner workings or expand what is there.

 

[KiwiMace]

Octave might be an easier step. Matlab-like and free.

 

[Enginerd9]

Free is good. Open source is also excellent for what I want to do. In the end, I plan to resurrect my old GA for shape optimization. It'd be nice to figure out how to get that to talk to whatever package I end up running.

I'm pretty confident I can get this Excel thing to work. It may seem silly, but for some fun side project, the fun is the challenge. I guess I should get my old class notes out.

Thanks!

 

[cowski]

I have found a nice solver algorithm package as a good starting point

Is the solver package functional code? If so, you may be able to get VBA to pass information between Excel and the solver. Reference thread770-279111 to see an example from Excel expert (and eng-tips member) Doug Jenkins (aka IDS).

 

[Enginerd9]

Yes! Great link!

http://www.basicnumerics.com/bflow/

It's actually a really nice package to start with. I've only run a couple of example problems through it, but it seems solid enough. I'm still evaluating it.

Thanks!