[waynezw0618]

[waynezw0618]

b absolute velocity (at the same conditon) calculated by CFX

[waynezw0618]

[waynezw0618]

Hi
i am sorry.too too much try maybe.and also there is no velocity(U=0)in hub and shroud boundays in my result.but i think there should be angular velocity like mixvessl2d'rotor.
i do not know why,can you help me ?

[dmoroian]

I agree something is strange in picture a and b. It looks like you are rotating the impeller in opposite direction to the one presented from cfx. However, what you show in a and b cannot be relative velocities since they are not along the blades (as seen in the first picture from cfx).

Dragos

[waynezw0618]

i agree with " they are not along the blades",and they do not normal to the blade too.
and also there is velocity in the hub and the shroud (which is up and down boundary of the impeller).

what is wrong with it?
wayne

[waynezw0618]

and also there is velocity in the hub and the shroud (which is up and down boundary of the impeller)
--->
and also there is no velocity in the hub and the shroud (which is up and down boundary of the impeller)

sorry.

[waynezw0618]

and look at next two picture (1 is foam ,2 is CFX)
1

2

you can find in foam calculation before the flow goes in to the blade it does not rotate,which is quiet differet from CFX result.
it also happened in the inlet boundary
so i guess it may be because of the faceSet or boundary condition,i am not quiet sure.
what is your idea?
wayne

[waynezw0618]

foam

[waynezw0618]

cfx

[dmoroian]

I still think the impeller in foam computation spins in opposite direction to the one in cfx.
foam: +0z
cfx: -0z

you can find in foam calculation before the flow goes in to the blade it does not rotate,which is quiet differet from CFX result.

Most likely this is due to the fact that you did not impose rotation on the hub.

Dragos

[waynezw0618]

Hi

I still think the impeller in foam computation spins in opposite direction to the one in cfx.
foam: +0z
cfx: -0z

i have checked that,the direction of rotation is the same in foam and cfx.or please look at the last two images,especially in cfx,the velocity vectors around axis (central of the blue circle in the underside of the cfx image,where the inlet pipe is nearby)are anticlockwise,if i impose a opposite direction,there would be an clockwise rotation for it near the inlet pipe,the roatation dirction of fluid in the inlet pipe and nearby will not intensively affect by the downstream.
so any other reasons?
_________________________________________________

Most likely this is due to the fact that you did not impose rotation on the hub.

Q2: how to impose the rotation on the hub? for the hub is wall,and MRFZones has boundary correction function which has impose the rotation to the boundary,and what will i do ? faceset(add hub and shroud to faceset or not delelet the boundary as what in the mixvessl2D?

[waynezw0618]

by last, i mean in cfx result, there is a rotation around the axia in evidence,the rotation direction is the same as i have imposed,but there is no rotation in Foam result.
i think it will a rotation of opposite diretion if i have spined a oppsite one

wayne

[dmoroian]

Can you display the MRFZones dictionary here?

[waynezw0618]

Hi
i am not working with my computer now.and i remember the omega is defined as [0 0 -1] 75.88
(impeller`s rotation speed is 725rpm, direction in -z).you could look at the "cfx" image carefully.at the right corner,z is to the inner,and the vector arround the inlet is clockwise.and angular velocity is in the -z.

wayne

[dmoroian]

This is why I want to see your dictionary, because my impression is that you have a rotation axis defined as

axis axis [0 0 0 0 0 0 0] (0 0 1);

in foam. So please post it when you will be able to.

[waynezw0618]

Hi
i am sorry for i am away for the passed few days. and now i am back to my office.
i have check for that right now,and you are right, i gave a wrong rotation axis dirction.i will modified it.
by how to impose the rotation to the boundary? do i need a faceSet?or others?

[dmoroian]

No you don't have to impose the velocity on the rotating wall. You just specify a normal wall "fixedValue" condition. If it is inside the rotating zone, then it will now how to impose the velocity correctly.

I'm glad you found the error,
Dragos

[waynezw0618]

Hi
how to impose?
i try to make a faceset like:

name IMPELLLER;//name of the cellzone
action new;
topoSetSources
(
boundaryToFace
{
}
);

and then modify the MRFZones Dictionary as:

5 //number of my rotating boundary..
（
IMPELLER //my cellZone
{
patches (HUB);//HUB is one of the rotating zone
origin origin [0 1 0 0 0 0 0] (0 0 0);
axis axis [0 0 0 0 0 0 0] (0 0 -1);
omega omega [0 0 -1 0 0 0 0] 75.92;
}
....//other rotating boundary such as shroud blade and so on
)

and specify all normal wall "fixedValue"

is that right?

wayne

[waynezw0618]

and here is the vector of U on "HUB" whichi is rotating