Hello

I'm a university student and have very little experience using fluent and wondered if any one could help me. I'm trying to model a turbine blade in gambit and then solve it in fluent for heat transfer. Could anyone tell me anything they think I should be considering or need to know before I can do this?

Any help would be GREATLY accepted

Thanks in advance

Nathan

hi Nathan,

I'm doing the same simulation. What do you need?

dan

Hi dan

Thanks for your response. To be honest I'm not 100% sure how to go about this problem. My project supervisor just doesn't seem to want to know. We don't have GTubro at my uni so I've been trying to model a turbine blade in gambit without it but have come up with something that looks more like an aerofoil. I looked but couldn't find any geometry I could use from the net. I'm about to start with Fluent but not sure what boundary conditions etc i should be using.

Thanks

Nathan

R u simulating a 2d or 3d flow ,i guess it should be a 2d simulation but anyways let me know this first

regards pratik

Hi Nathan, I don't think you can properly model a turbine blade (expecially in 3D) without the correct geometry information. If you model a blade passage you can use Pressure Inlet and Outlet B.C., Periodic on side boundaries, and Wall B.C. on pressure and suction sides of the blade.

hope it helps

hi pratik

thanks for your response. It will be a 2d simulation.

Cheers

Nathan

ok dude i will help u out , since its time out for me , i will check u 2morrow , ok

ciao pratik

hi Thanks for your help. I've modeled a 2d shape to resemble a turbine blade. Not sure what else i can do really without Gturbo. I'm about to start using it with fluent. Should i be using a single blade or using two and observing the flow between them. Thanks again Nathan

Nathan, you can simulate a single blade passage (one blade) and apply periodic b.c. midway into the passage to simulate an infinite blade row.

I see what you mean now. That's great thanks . Going to give it a go with Fuent tomorrow. Is there anything else i'll need to know before i can simulate a flow? nathan

ok mate, thanks alot. nathan

well, i think the suggestoins from dan seems to be the way u must carry out the simulation , i have a question , are you also loookin for heat transfer or just flow simulation? ,have u decided how big will be your domain upstream and downstream, .if not ,please look into this factor, generally the upstream should be around 6*Diameter of the entrance , while downstream could be same or more .

U must anyways check it with the convectional figure as well. Using periodic boundary condition is correct , or even symmetry be used ,i guess as u r not excepting the flow normal to the boundary which u will name symmetry or periodic . either of them should be fine , what do u think dan ?

try to use a quad pave mesh it may help u to get good results

anyways i hope this should help , if u need some more help kindly let me know

regards pratik

I think that is ok. You can also refer to the chord length of the blade to place inlet and outlet of the domain, place the outlet far enough to avoid reversed flow messages.Quad mesh is the best choice if the flow is aligned with the boundaries, but triangular mesh may be easier to create. Also, create a b.l. mesh on the blade and check the y+ at the end of the simulation.

regards

i agree , so nathan do u have any update from your simulations?