Learning Forum

[mohammedlaminemekhalfia]

Dear ANSYS Users.

I am using ansys fluent CFX, I would like to have your help in resolving the problem mentionned in the attached Picture.

according to your experience, What would be the mesh quality and the good meshing practice that I need to follow. I tried a lot to have an idea about it through reading a lot of article, yet I couldn't guether a good idea about it.

for the Boundary condition, wall nomination, is the structure correct?

I am using CFX, is the result in fluent will be different or more accurate?

[Rob]

If you're trying to replicate the experiment the domain is far too small: where does the air go in reality compared to where if goes in the CFD model. Both codes will handle the FSI in this case: the benefit with the recent versions of Fluent is the intrinsic FSI may be easier to use than system coupling. n

[mohammedlaminemekhalfia]

I used a small domain since I am assuming that dissipation is too small and that there is a total transfer in enery from the Nozzle output to the Blade surface.nalso I ignored air range in reality because I ( in my thought ) that any registred effect on the blade would be because of the direct impact of air in principle.nI do not know if the logic is correct or not that is why I post here so I would benifit from wonderful minde like youand allow me to thank you for being so helpfuln

[Rob]

You're welcome. If you look at where the air is going in reality compared to the simulation, would you expect the force to be the same? Ie what is the flow like passed the blade?n

[mohammedlaminemekhalfia]

I think yes, if you can look to the picture the face that is to the X direction is the output yet I will generate the solution for further X distances of the output.nI built that idea since I believe that the vibration will be more due to the direct impact of air pressure. not the variation of environmental pressure since it still too low. we also can't consider this excitation as a wind since it take place on vertex not a total face. by the way is it is first stage gas turbine blade and for the moment it is static later I will perform the same experiement in a chamber. so expect a lot of questions about in the future nalso I would like to have you advices about mesh methods and good meshing practice, for me I am sure I am doing that in a wrong way, I am just let the solver generate mesh automatically I know that it is not always correct but I am still having a big problem in meshing although I read a lot about but still sufferringnnn

[Rob]

The AIC courses will give you some details on meshing as will the tutorials. The mesh is there to capture the flow, so whilst y+ is a good way to estimate the first cell height it isn't much use in checking if there is sufficient mesh to resolve the flow. nRe the domain. Flow hits the blade and is deflected sideways. In reality flow won't straighten up again as there's no reason to. In the CFD model it hits the domain side, turns and then (probably) interacts with the blade again. For the full model you need to read up on periodic meshes. n

[mohammedlaminemekhalfia]

in reality it depend on the position of the blade to the Nozzle and the ratio between the Nozzle output section to the blade section.nYou are true in reality flow won't straighten up again, the CFX results show that two the air dessipate after hiting the blade. so I am in the right way.ncould you please send me the Link for the AIC courses unfortunately since a week I can't enter courses section in the ansys forum.n---- you mean by periodic meshes to transform my domain to sub-domains and submit different size and methods on each subdomains?.thanks again

[Rob]

AIC courses are at the top of the page, no need for a link. If you can't get to them clear the cache etc as they should be available for all. nPeriodic meshes are used to represent longer or cyclic geometries, in your case you can model a single blade and then mimic the effect of a full turbine fan. n

[mohammedlaminemekhalfia]

I am still in the case of one single static blade that face air excitation.nI exported the pressure to static structural to view Blade displacement. since static structural show only the amplitude of displacement due to excitation.nnI would like to know how can I plot vibration at the tip of the blade since there is no frequency of excitation nor frequency of rotation so I can not go to harmonic? even transient is it also adequate for this case?n

[Rob]

You may need a transient FSI for that, can advise. n

[Stephen Orlando]

for a 2-way FSI transient simulation, please see this tutorial: https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v211/en/sysc_tut/sysc_tut_oscplate_export_cfx.htmln

[mohammedlaminemekhalfia]

Sorry but the link require a special accountnn

[Rob]

https://forum.ansys.com/discussion/22563/using-help-with-linksn

[mohammedlaminemekhalfia]

Array Array I will go through the tutorial and I will return to you once I have difficulties as usual , Thanks a lot for both of you.nArray please for one way coupling how could the plot this curves?n n