CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Phoenics (http://www.cfd-online.com/Forums/phoenics/)
-   -   Re: how to do this turbine blade passage flow nanl (http://www.cfd-online.com/Forums/phoenics/51997-re-how-do-turbine-blade-passage-flow-nanl.html)

 manish January 12, 2006 12:37

Re: how to do this turbine blade passage flow nanl

hi to all .I want t osiumulate flow through bladepassage of a radial inflow gas turbine.It has presure drop of 2.8,so i have not to use stator.I have 12 no of blades.I am taking only one.It has w=1680 rad/sec (60,000 rpm).Inlet pressure is 2.83 bar (Total pr) outlet is 1.12 bar.Inlet total temp is 923 K and outlet temp is 760 K.mass flow rate is 0.596 kg/sec.Flow velocity at inlet 125 m/sec.I have given different boudary conditions but at each time it is giving very erronous results.i.e .pressure goes upto 9 bar.

(1) when i a m giving mass flow arte at inlet and pressuire at outlet then it is giving erromous resuts.mass flow is conserved but pressure values are far from actualk valus. (2) AT INLET Pressure and oulet pressure .Here pressure valuse have less difrence wit hactual values but mass flow is not mathich. can anybody guide me in this matters.whic hboundary condition i should give.and which initial valus i should give? . second thing is that can i go directly by giving unsteady anal;ysis .or first i have to solve this problem using steady satte and then go throgu hunsteady state. Thnks all.

 Mick February 21, 2006 10:59

Re: how to do this turbine blade passage flow nanl

These problems are not easy.

You should definitly get a steady solution first, otherwise an unsteady solution will not be possible.

This huge error is very likely due to the solution being bad, do the vectors look wild and unrealistic.

If your case is BFC.

Problems may be because of your mesh or possibly because of a possible anomaly with BFC xcycle that I think I have noticed. If so this unfortunately is not an extremely quick thing to turn over a fix for.

An alternative approach (but somewhat expensive computationally)could be to use MOFOR. If you have or can make an STL file of your complete fan geometry. I can find a similar type example which could be used by you as a set-up guide if you are interested in this option.

 All times are GMT -4. The time now is 04:06.