Problem in the definition of dynamic mesh zones


I have a simple channel with length = 2 mm, Height = 150 um. Inlet is a mass-flow inlet and the outlet is a pressure outlet. The liquid is water and the flow is laminar. The bottomwall (see picture) need to move up and down (along y) in order to generate pressure waves that propagate in y direction. The movement of the bottomwall mimics the effect of a piezoelectric transducer.

My problem is that when the bottomwall moves, a perturbation origins at the outlet maybe for a conflict between the pressure condition that is fixed and the variation in pressure of the waves.

So I thought to move only a part of the bottomwall (figure below, black line stand for moving part of the wall, blue line represents the part of the wall where the condition of dynamic mesh is not applied), but in this way I have a jump in the condition applied to the bottomwall: on the moving part I set dynamic mesh (rigid body), on the other part I don't know how to define it. (If i put stationary, I have a perturbation).

Any advice?


  • Glad it has moved. In reality, if the wall moves, it will cause this perturbation. So, I believe one option you have is to decrease this fixed length .. second option is to make this part as porus moving wall. But both options are not a real simulation of what would happen if the wall moves

  • Sorry @YasserSelima, I didn't reply anymore. I used a profile instead of a udf, and it worked.

    1) What do you refer with decrease the fixed length? Why should be beneficial?

    The only thing that I did is to have all the bottomwall that is moving, and increase the lenght of the channel , moving away the outlet, but it is not a real solution.

  • Hello,

    Can you not extend your channel length so the outlet condition is much farther from the moving wall?


  • Hello @Kremella ,

    Maybe I generate bit of confusion: in the first case bottomwall extend for all the length of the channel. In this case I have a perturbation that originates from the outlet.

    In the second case i divide the edge "bottomwall" in two parts. The first part ( black line in the second image of my first post) is moving and I applied a dynamic mesh condition. The second part (blue line in the second image of my first post) is not moving and I didn't apply a dynamic mesh condition on it. But a perturbation originates in the point where the different conditions (moving and not moving) are applied.

    The problem is that the perturbation propagates in the channel (on the left).

  • @Luigi0 Try to think about it. A perturbing wall causes pressure perturbation. This happens in reality .. If you want to suppress these perturbation in reality, you might need to a screen or honeycomb. Or you can make large chamber at the end of the perturbing wall. You can do any of these in your geometry as well. But is this the real case you want to simulate? this is the most important question ...

  • I want to simulate an acoustophoretic device. I concentrate on the part where the piezoelectric transducer vibrates generating acoustic plane waves in the channel, and the goal is to trap particles in some position in the channel ( the frequency value is chosen to have a pressure node at h\2, along height). In the real case the channel is longer and the actuation is present only in a specific part of it.

  • so, why don't you make a large expansion at the end of the wall, like a reservoir. I am not acoustics guy, but I think this will be like an open channel. then you can move the whole wall without caring of what happens in the reservoir.

Sign In or Register to comment.