|
[Sponsors] |
December 9, 2010, 10:38 |
Bachelors Project Using FLUENT
|
#1 |
New Member
kvrmurthy@ovi.com
Join Date: Dec 2010
Posts: 4
Rep Power: 15 |
Hello forum,
As the title suggests i am planning to do project in fludization using FLUENT. As i am a beginner i want to know some information regarding the versions and the available documentation as well as the books to learn FLUENT. I came across FLUENT 6.3 and the latest ANSYS Suite 13 which has FLUENT. Which one is best suited for general use in situations like this? Is there documentation or books i can read to learn this software? Thanks, your replies are welcome. |
|
December 9, 2010, 19:25 |
|
#2 |
New Member
Oscar
Join Date: Jun 2010
Location: Sheffield, UK
Posts: 18
Rep Power: 15 |
Hi and welcome, I'm also new in CFD world, I've been working in a few projects in my school and in a research center (in Mexico), and just like you I make my bachelors thesis simulating a solar water heater with Fluent 6.3.26 and was a lot of work but lets you learn a lot.
As a rookie I can't distinguish differences between the functionallity of Fluent 6.3.26 and Fluent 12 (included in ANSYS 12 suite), but ANSYS 12 Workbench allows you to create all the analysis in a single file, and in a single software (including CAD, mesh generation and processing the case), BUT (here's the big deal) there are some options that ANSYS Workbench that are missed specially for mesh generation (so I had need to use ANSYS ICEM CFD). I suppose that you love to be all day in front of the computer drawing first, meshing the computational domain (probably several times) and then taking care of the simulation process (nursering, like a crying baby), so probably you need a soft chair and a cup of cofee, specially if you're using the school's computers like me. About the literature, first I read all Versteeg-Malalasekera book (It's essential to know how the CFD works theoretically) and then I review all the models that involves the simulation and studying all of them using Fluent Users Guide and if the information are confusing I turn to Thermodynamics, Heat Transfer, Fluid Mechanics and Fluid Dynamics books (you should know of your previous courses), yeah, it's a lot for read (a kindle must help). As advice try to simplify the computational domain, because a big domain will need a lot of computational power, create the best mesh that you can, then assign the correct boundary conditions and not despair if you obtain a lot of errors sometimes are normal, check them and solve it. You can post it here in the forum and for sure somebody may help you. At the end you've gained a lot of knowledge and probably you must fall in love with CFD an turn to your MsC, PhD or what you want . See you. |
|
December 9, 2010, 20:33 |
|
#3 | |
New Member
kvrmurthy@ovi.com
Join Date: Dec 2010
Posts: 4
Rep Power: 15 |
Quote:
Thanks a lot, Bye! |
||
December 9, 2010, 20:40 |
|
#4 |
Member
Tiago Macarios
Join Date: Mar 2009
Posts: 35
Rep Power: 17 |
fluent 6.3 and fluent 13 are very much like. Just some very specific modules have been updated. You can have a look at the ansys customer center to see if any module that you are interested has changed.
If I were you I would go for fluent 13 it has all the same capabilities of 6.3, just mind the bugs... |
|
December 10, 2010, 11:27 |
|
#5 |
New Member
Oscar
Join Date: Jun 2010
Location: Sheffield, UK
Posts: 18
Rep Power: 15 |
Well ANSYS 12 (the version that I use) don't read .msh files of mesh generated by ANSYS ICEM CFD or Gambit, and that's become a big trouble, specially if you need an specific mesh (hexa mesh or something) but if your geometry isn't complicated i suppose its better ANSYS Workbench suite.
|
|
December 10, 2010, 12:58 |
|
#6 | |
New Member
kvrmurthy@ovi.com
Join Date: Dec 2010
Posts: 4
Rep Power: 15 |
Quote:
|
||
December 10, 2010, 13:02 |
|
#7 |
Member
john
Join Date: Aug 2010
Location: pasadena, va, usa
Posts: 36
Rep Power: 15 |
i found the following tutorials useful when i first started working with FLUENT:
https://confluence.cornell.edu/displ...arning+Modules Unfortunately, the meshing is done with GAMBIT. However, you can download the ready-made meshes from the site, or... you could use it as a good opportunity to learn ICEM CFD. Good Luck, john |
|
January 27, 2011, 10:34 |
|
#8 |
New Member
Dan Adrian
Join Date: Jan 2011
Posts: 12
Rep Power: 15 |
Hello! OzMantle, I have a similar project as you did, only I have to study the heat transfer inside a flat solar panel (http://www.igneus.ie/images/flat_plate_cut_away.jpg), and I wanna know if its possible to simulate that in one mesh (3D), representing the hole body of the solar panel, that would incorporate all components, and off course taking in considerations all the heat transfer that intervene( convection, radiation ....). What I have managed so far, is just to simulate the heat transfer from the pipes to the fluid inside them. But I cannot manage the simulate the heat transfer from the absorber plate to the pipes and air around the pipes, and I also don't know how to set the correct radiation parameters to heat up the absorber plate according to the radiation intensity and the external temperature.
I hope you can give me some tips in doing this, I need it to finish my bachelor project whit Thank you in advance. |
|
January 27, 2011, 15:46 |
|
#9 | |
New Member
Oscar
Join Date: Jun 2010
Location: Sheffield, UK
Posts: 18
Rep Power: 15 |
Quote:
If you're trying to simulate the whole system you may need a lot of computational power (cluster, gpu processing, or something), CAD work seems really easy, but when you turn into CFD some troubles arises, first the meshing process can take a lot of time (for a good meshing) because you will have a lot of bodies in the entire part, each body needs a mesh and probably those real geometry are difficult to capture, then you need to control several variables on the simulation process, for the radiation parameters FLUENT has a calculator referenced of your location. Note that you must encapsulate the system into one body wich represents the exterior (air surrounding). Conclusion: It's not impossible but requires a lot of work and alot of resources, for me was impossible to do a simulation of the entire system (I'm from Mexico, public school and I don't have $ ) so I start reducing the complexity of the analysis finally I simulate only the fluid body, if you decide to do the analysis CFD-online it's a good forum so here you can find a lot of help or the advice that you need. If I can help you, just ask. We keep in touch. OzMantle |
||
January 27, 2011, 16:09 |
|
#10 |
New Member
Dan Adrian
Join Date: Jan 2011
Posts: 12
Rep Power: 15 |
Thanks for replying
I know simulating the hole system takes resources, but I think my PC might take it, or I at least I will find out eventually :P. Anyway, the reason I want to take in consideration the hole system, instead of concentrating on each component at a time, is because (in my opinion), I would obtain results more close to reality. But then again I'm a novice in CFD applications, and I might be wrong. So far I managed to create the mesh ( I am using Gambit, I tryed to export from CAD but I couldn't :| ) for the entire system, using Tgrid scale for volumes, and the grid got exported whit no problems in Fluent. But here is where I encounter my main issues. I didn't knew until you told me that I have to created a block that represents the ambient air, which might be a key component in this. Also I am not sure what types should I declare the to plate that are above the actual pipes ( first plate is a transparent glass, the second is a copper plate whit high absorption proprieties) should I declare these 2 as "wall" or "velocity inlets"? For now these are the things I need to know, In case any other issues appear I will make the public here :P Thank you again, and we will keep in touch |
|
January 31, 2011, 11:24 |
|
#11 |
New Member
Dan Adrian
Join Date: Jan 2011
Posts: 12
Rep Power: 15 |
And another thing I wanna ask you is, how to transfer the heat (by conduction), from the copper absorber plate, to the pipes ( also made out of copper) which are positioned under the plate. The plate and the pipes are welded together.
|
|
February 3, 2011, 02:07 |
|
#12 |
New Member
Dan Adrian
Join Date: Jan 2011
Posts: 12
Rep Power: 15 |
I'm using DO radiation model, whit solar load, on the copper plate inside a insulated box, whit the top wall being a semi-transparent glass, and all my initial temperatures are set to a minimum of 278.15 K. Why is it , after I iterate I get a temperature range from 232 to 278 K?
My radiation is set correctly to 1000 W/m^2, why do I get the sub zero temperatures? |
|
December 17, 2011, 02:39 |
This may be a little late, but I am working on a similar project
|
#13 |
New Member
Steve Schwade
Join Date: Dec 2011
Posts: 1
Rep Power: 0 |
I used a series of steady state control volume analyses’ to find the heat flux into the heat exchanger as a function of the collector surface temperature and the incident beam radiation intensity. I am not new to FEA, but I am new to CFD – so any advice you have would be much appreciated. If you want to used the results in your analysis, just let me know and I can give you all the assumptions/constraints and all you would have to do is define the surface heat flux in your simulation as a function of temperature (or time if you feel like making an initial guess and iterating then manually imputing the values on a time basis a few times.
https://docs.google.com/open?id=0Byt...JkMDNlOGQxN2Y5 https://docs.google.com/open?id=0Byt...JiYWRmM2U1ZGVm |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
How to open Icem mesh in Ansys Fluent? | emmkell | FLUENT | 27 | February 6, 2018 03:34 |
Fluent 6.3 32bit vs Fluent 12.0 64bit | ibex7 | FLUENT | 7 | April 18, 2011 02:44 |
Fluent 12.0 is worst then Fluent 6.2 | herntan | FLUENT | 5 | December 14, 2009 02:57 |
Problems in lauching FLUENT | Lourival | FLUENT | 3 | January 16, 2008 16:48 |
Advanced Turbulence Modeling in Fluent, Realizable k-epsilon Model | Jonas Larsson | FLUENT | 5 | March 13, 2000 03:27 |