CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Regerative pump backflow/ reverseflow problem (https://www.cfd-online.com/Forums/fluent/44273-regerative-pump-backflow-reverseflow-problem.html)

Ravindra March 27, 2007 23:06

Regerative pump backflow/ reverseflow problem
 
I am working on regenerative pump. I have tried for all B.C. i.e velocity inlet and outflow , mass inlet and outflow , pressure inlet and pressure outflow. For pressure inlet and outflow condition at least I can see the development of pressure contours but other cases results are not proper. For pressure inlet and pressure outlet condition I am giving 1 atm pressure at inlet and at outlet 9.6 atm which is pumps calculated head. I just want to see how the flow pattern is there in the pump. But there is total reverse flow in the pump. But the impeller action is visible after seeing the velocity vectors. To reduce backflow problem I have increased area by connecting pipes at the both inlet and outlet ends . I also reduced Under realization factors. I tried URF for momentum as 0.1 , 0.001 similarly I reduced other URF also. But no effect is visible. I have given 2900 rpm rotation to moving parts of the pump that is impeller , blades etc. and no rotation to fluid domain. Similarly I tried with the case moving walls with no rpm to moving parts and fluid domain moving with the 2900 rpm. Please help me . How to minimize the backflow in my case?


aja March 28, 2007 01:43

Re: Regerative pump backflow/ reverseflow problem
 
Check the parameters of the pump, if there are near the expected optimum of pump performance. Otherwise, try to decrease (or increase) pressure difference at the pump. In case of the radial pump impeller, there is a large operation area, where the back flow at the impeller outlet exists. If your parameters are far from best efficiency point, you will obtain back flow at each combination of inlet-outlet. If tour design and operation point is O.K., the solution should to use Euler f.m.g. initialisation, or behaviur could be improved by apprpriate back flow velocity vector direction at the outlet. Or if you analyse impeller alone, try to extend domain at the impeller outlet.

Ravindra March 28, 2007 02:16

Re: Regerative pump backflow/ reverseflow problem
 
thanks i have used FMG init when Under realization factors are reduced but the thing is that as now i am giving resudal levels 10 power -3 in this case i am getting very fast covergence with in 140 iterations but URF is reduced then it require more iter in that case i am using fmg Another thing is that pressure differnce reduction means what i have to do? Currently i am using 9.6 atm pressure at outlet and at inlet i am using 1 atm pressure should i reduce at the outlet or should i check for starting with 3atm - 9 atm or should i give target mass flow rate at that outlet and pressure zero condition and at inlet some higher pressure? plz tell me


aja March 29, 2007 03:43

Re: Regerative pump backflow/ reverseflow problem
 
I must know the design parameters of your pump. If at your parameters the flow is not very nice, try to shift pressure difference head( in case of pressure inlet ), or mass flow ( in case of velocity inlet ). Use for example two lower and two higher values for inlet b.c. and look at the results. You can use FMG init everytime and URF are better lower.

Ravindra March 29, 2007 05:02

Re: Regerative pump backflow/ reverseflow problem
 
Hi thanks currently i used velocity inlet and pressure outlet condition where i have given both target mass flow rate as will as pressure 9.6 atm which is head of pump and i have reduced URF drastically like momentum factor 0.005 , pressure 0.01 and other factors also reduced slightly so now i am getting some what proper velocity vectors but now though pressure is showing proper contour values but development of pressure is not clearly visible and swirl inside the pump is completely lost ( i have used K-epsilon model). so what should i do to see proper velocity vectors ( with minimum backflow) , devlopment of pressure contours and swirl inside the pump.



All times are GMT -4. The time now is 07:46.