I am using mass flow, normal to boundary for the inlet; and total pressure, circumferential, and pressure set to zero. The pump is power by a rotating impeller, and there should be a pressure raise; but the result in CFX post show a pressure drop from 50kpa at the inlet to 0 at the outlet. How is it so?

In additional, how do I measure the pressure change along the pump, and plot the pressure change along the blade?

Please help. Any suggestion is welcome

Dear Richard,

Are you using a rotating domain? If so, are you using Thermal or Total Energy for the heat transfer model? Only the Total Energy model will account for work input due to the rotating impeller, and you will see the pressure raise you expect..

If you are already doing the above, I do not know what else may be wrong...

Good luck, Opaque..

When you talk about pressure drop, are 50kpa and 0pa static or total pressure? Are they relative pressure? How far inlet and outlet are away from blade?

Not sure I am with you on this....Opaque

If the flow is incompressible, which it usually is for a pump, you do not even need to solve an energy equation.

The only need for any energy equation is if the fluid properties vary strongly with temperature. For example if there is significant viscous dissipation which changes the fluid temperature and this changes the properties (density or viscosity), if the fluid is an ideal gas, etc.

If the domain is rotating, you want to make sure you look at absolute frame pressure changes across the pump!

Regards,

Bak_Flow

Hi Richard,

It is entirely possible for the static pressure to drop across the pump. The pump will raise the total pressure, but the local static pressure will be a function of the velocity as well. Keep in mind that Ptotal = Pdynamic + Pstatic.

Check the change in total pressure across the pump. If you still see a drop, look at your flow angles at the inlet. A common mistake is to rotate the domain in the wrong direction or get the wrong inflow direction at the inlet. If you specified the inlet in the rotating frame, normal to boundary will give it pre-swirl. What you may have intended is normal to the boundary in the "stationary" frame.

Regards, Robin

Hi all, I think many people are working on axial machines.I just started my work on the same.Can you please suggest me how can i make my flow domain in ProEngineer for an axial pump or some resources to make it?. Thanks in advance.

Manu

Richard,

in my earlier post I was incomplete: make sure you are looking at absolute frame total pressures...static pressures are of course frame invariant.

In this regard you should see a rise in the absolute frame total pressures. Losses in a rotating passage are seen as a loss in the rotating frame total pressure.

Regards,

Bak_Flow

Thank for all the reply

There are 5 domain of this simulation. The inlet, straightener, impeller, diffuser and outlet. All are stationary domain except the impeller.The flow inlet length is about 160mm and the flow outlet is twice that length. Along the impeller, i should expect some pressure gen due to the rotation, but instead, i get a pressure drop. The pressure differences of the inlet and outlet are total pressure in stationary frame. However, there is pressure gain along the straighener, which i also do not understand.

About the total enery model, how do i apply it? I willingly to give it a shot.

About the rotation direction, we could only select rotation and stationary domain, but cannot select clockwise or anti clockwise in CFX pre, so how do we control the rotation direction

All suggestions welcome

Hi Richard,

The positive rotation direction is based on the right hand rule. If your pump rotates the other way, enter a negative value.

Regards, Robin

Hi

We can set direction of rotation by right hand rule.

Sachin

thank for the reply.

After solving this issue another comes up, I am using Ansys CFX 5.7.1. I could no obtain a convergence, hence I suspect it is due to the meshing. I need to increase the mesh. what type of meshing should i use for the blade and the interface?