a simulation with rotation and vibration
 

description of the problem: the fluent flow to the hole with the pressure 0.8MPa along the tool which is the capillary. this is 1/4 the geometry model in the figure. the tool rotate and vibrate with high frequency.
i do the simulation about rotation and vibration respectively. when the rotation simulation is done ,the tool model is set the immersed solid and when the vibration simulation is done, i just use the fluid model and set the mesh deformation. all the simulation ara sucessful to compute. the fluctuation of pressure and velocity appear.
questions:
1. Are these simulation right?
2.when i want add the voltage to the tool and the workpiece and do the simulation about rotation and vibration of tool simultaneously ,how to do? the immersed solid do not support the electric field model. when i set the tool as solid demain, and the tool set the rotation, the error appear.
3. when add the vibration to the tool, how to do? add the mesh deformation to the tool and fluid or and the mesh deformation to the fluid and set the solid motion equation to the tool. when the solid motion is set, the electric field model is not used.

Attachment 24939
there is the geometry model.

1) I have no idea if your simulation is right. I do not understand what you are modelling.
2) What does the voltage do? Why do you want to model it?
3) If the vibration is known in advance then use moving mesh to implement it. Do not use rigid bodies for this.

Thank you for your reply.
The fluid is brine which contain water, NaCl and gas. when there is voltage on the tool and workpiece, there is current along the workpiece ,brine to the tool. I want to simulate the change of the characteristic of the brine when the tool rotate and vibrate, such as conductivity , temperature , gas volume fraction distribution ,pressure and velocity and so on. i don't know is the model right. Maybe just need the fluid model to simulate this problem, set the rotation and vibration on the interface of the tool and brine, set the voltage on the interfaces of the tool and brine , interface of workpiece and brine.
the vibration equation is known, i can try the moving mesh. thank you very much for your advice.

I do the simulation about rotation and vibration respectively.
About the rotation simulation
one way setting the tool domain as solid domain and another way setting the tool domain as immersed solid,which one is right?
About the vibration simulation
one way setting the tool domain as immersed solid and the motion as specified displacement,give the Z direction a displacement quation. the result show the fluctuation of pressure but too small. Another way just using the fluid domain. setting the fluid domain as mesh deformation,and gaining the mesh displacement: areaAve( Mesh Displacement Z)@wall and the velocity: areaAve( Mesh Displacement Z)@wall. the result show the fluctuation of pressure and it is bigger.i want to konw which way is right?

I do not understand your question, but you describe a number of methods to model the motion (immersed solid, moving mesh) and they are all useful for some cases. But whether it is appropriate for this simulation I cannot say as I have no idea what you are modelling.

the case of the simulation is that :
the tool is made of stainless steel, which connected the negative of the power supply. the motion of the tool is rotation and vibration.it is capillary tube. the work is made of stainless steel,which connect the anode of the power supply. it is stationary. the brine which is conductive electrolyte flow along the capillary to the work. the current flow from the work , brine to the tool .so is the simulation.
i have done the rotation and vibration respectively. but there is no voltage in the simulation. for this time, the voltage is added to the simulation. is it of coupling problem. if it is the coupling problem, it is not estabished on the ANSYS WORKBENCH. the electric uint and cfx uint can not be coupled.
the simulation is done by the ANSYS WORKBENCH using the coupling method and by CFX using the electromagntic model, but both are failure. i have no idea how to simulate? maybe the fluent ?

What type of vibration? In which direction? Is it coupled to the flow or imposed?

the displacement equation of the vibration is sine function along the axial direction of the capillary tube named Z direction. it is imposed to the flow by ultrasonic vibration.

How is the electric component coupled to the flow? If the fluid is a constant resistivity then the current flow will just be related to the tube position and there is no need for CFD.

if there is gas in the fluid, the resistivity is different . and the temperature effect the ion activity,so the resistivity of the brine is not the constant.
such as caviation ,if the pressure is lower, the gas is produce from the fliud. when the tool is ultrasonic vibration, the low pressure will appear. i think it worth to simulate.

This is looking like a very complex model. Have you done simulations where you simulate these effects by themselves - that is a cavitation only model, a moving mesh only model, and an electric field only model? Do not attempt the model with everything before you can successfully simulate the physics in isolation.

i have done the rotation simulation with cavitation model and moving mesh with cavitation. but the rotation simulation and moving mesh simulation can not support the electromagnetic model.so the electri field can not be simulated . i am doing this now .i have try a lot ,but failure. so i need some idea !

Is the electric field model just to get the current flowing from anode to cathode? In other words, there is just a variable resistance in the fluid section, and you only care about the current flow? Or does the electric field do more than that?

yes. i need just the current. the gas volume fraction effect the conductive.

So if it does not want to do this using the maxwell equation solver, can't you put a simple diffusive additional variable on to represent electric potential with the diffusivity set as the conductivity of the material? You will just have to transfer the units to diffusion.

Thank you for your reply. i have done a simulation with the voltage in the vibration simulation.I set two solid domain and a fluid domain. the voltage is added to the tool and workpiece respecitively. Does the cfx support the maxwell equation solver.

there is the electri field of a point in the fluid in every time. But i can not gain the current or current density. why

I have never used the maxwell solver in CFX so cannot help you with details. If the variable you are looking for does not appear to exist then create it as a user variable.

Your time step size looks horribly coarse judging by the facets on your time history. You are going to need MUCH finer time steps than that.

Thank you for your advice. in the simulation,when the mesh quality is fine, the error appear ,so the mesh is coarse. i will define small time step.

hi，ghorrocks. as the immersed solid can not used in the multiphase flow, so the rotation simulation is not correct in my simulation. i want to set the rotation simulation. but as the tool is set the solid domain and the fluid is set the fluid domain. as you konw the tool rotates in the simulation, i set the tool as rotation in the domain motion,but how to set the wall function. when i set the fluid domain as rotation, the wall volecity of the interface of the tool and the fliud is set counter rotating wall . but two kinds of simulation are not correct.
can you give me how to set the rotation just using the solid domain and fluid domain . a example is ok

please give me some advice!!

I have nothing to add beyond what I have already said. Until you consider my previous suggestions I see no point making more suggestions.

the rotation simulation have done with the single fluid,but when the fluid is set the multiphase flow, the mistake appear as follow:
Run mode: partitioning run

Host computer: MK-PC (PID:788)
Job started: Wed Sep 11 10:45:40 2013

Details of error:-
----------------
Error detected by routine PSHDIR
CDRNAM = ../../JCONT
CRESLT = NONE

Current Directory : /FLOW/ALGORITHM/ZN1/SYSTEM/EQUATIONS/JCONT_FL2/TERM1

ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Stopped in routine MEMERR

An error has occurred in cfx5solve: |
| |
| The ANSYS CFX partitioner exited with return code 1.

This error says you missed a fundamental part of the multiphase setup -but I cannot tell exactly which. I would check the multiphase setup again.

You could also post the output file with the CCL and error included.

Thank you for your reply. there is the output in the attachment.
the mistake is : No data source for Workspace, cannot monitor Solver data.

the rotation simulation is done when i set both of the symmetry area of the tool domain and fluid domain are domain interface and the rotational periodicity is set in the Interface Model. the connecting areas between tool domain and fluid domain, workpiece domain and fluid domain is set frozen rotor in the Frame change/Mixing Model. The electrimagntic model is set in the tool domain ,fluid domain and workpiece domain respectively. when the fluid contains only the brine,the simulation is finished without error, but when the gas (air at 298k) is contained in the fluid, the mistake appear such as above. i check the set in the simulation for many time, but i can not find the reason. Please tell why.
Anther question :
I do not know the really situation as the simulation, in my opinion，when the tool rotats in the fluid domain, that is to say it stir the fluid,so there must be change in the pressure and the velocity of the fluid. But i can not find the change in the simulation,so i can not believe the result of the simulation. Maybe the set of the domain interface is wrong,such as the Mass And Momentum, i try the Free Slip wall or No slip wall. but the result is no change.
Is the method for the rotation simulation right?

there is the pressure of monitor point in the fluid domain.

This is a very complex model and I do not have time to debug it for you.

All I can recommend is to build the simulation up piece by piece. Can you run a cavitation model without electrics, rotation and moving mesh? Once that is working, add the physics one bit at a time to build up to the full model.

Thank you for your advice. i will do the simulation piece by piece. but i can not find the reason for error. there is few books about the ANSYS cfx, can you give me same advice or others. my Email is lzx_527.student@sina.com

why ?
 

why is the result?
the rotation simulation is done and the tool domain is set as rotation with speed 600rpm
the explation of the pictures.
the model (picture 1): 1 is the interface of the tool domain and the fluid domain at the inside of the hole of the tool , the set:interface model: General Connection,Frezon Rotor; the interface side in fluid domain is set No slip wall, and the wall velocity is rotating wall :-600 rpm
2 is the interface of the tool domain and the fluid domain at the outside of the hole of the tool, the set is the same as 1;
4 is the interface of the tool domain and the fluid domain at the bottom of the domain. the set is the same as 1;
3 is the interface of the workpiece domain and the fluid at the side of the hole of the workpiece, the set interface model:General Connection,Frezon Rotor; the interface side in fluid domain is set No slip wall, but the wall velocity is set as 0 rpm;
5 is the interface of the workpiece domain and the fluid at the bottom of the hole of the workpiece, the set is the same as 3.
the quesetion is :
1. when the velocity contour is gained ,why is zero(the blue) for the velocity at the surrounding of the tool ;(the blue) in picture 2
2. when the velocity coutour is gained,there is no change for the velocity in the direction. in my opinion maybe is the swirl, and the direction of the velocity should be changed.
the tool: outside diameter 1.2mm, the inside diameter:0.8mm
the hole in the workpiece :diameter:1.6mm

is the set of the rotation wrong? or for there is no change for the too small tool in the fluid simulation.

other questions:
1.such as this kind of rotation simulation,the tool domain is set as fluid in many similar example. what is the difference between the fluid domain and the solid domain. the simulation when the tool domain is set as the fluid domain, the result is different, and the velocity surrounding the tool domain is not zero. why?
2.Is the CFX suitable for the simulation with two-phase flow and electromagnetic model? i try for many times, but failure.
Thank you very much!

I say again - This is a very complex model and I do not have time to debug it for you.

But I can see in your contours that it looks very blocky. This suggests you have a very coarse mesh. You will not get very useful results with such a coarse mesh, a far finer mesh is required.

Thank you for your reply. As the mesh is finer, the computer can not work. So, i just simulation with the coarse mesh. But i do not know the method for the simulation is right?

Then get a bigger computer. There is no point in doing such a complex simulation if you cannot resolve the flow. You are only wasting your time.

Yes. i will try!