Henrik Sönnerlind
COMSOL Employee
Please login with a confirmed email address before reporting spam
Posted:
10 months ago
Feb 1, 2024, 4:13 a.m. EST
I studied a similar problem about 25 years ago. It is a very interesting problem, because the apparent seal stiffness (and damping) becomes strongly dependent on the compression speed. By selecting an appropriate size and spacing of the evacuation holes, you can tune the dynamic properties of the seal.
An approach that worked well (experimentally verified) was to not model the evacuation hole, but just solve an extra equation.
The assumtion is then that the evacuation flow velocity is controlled by the difference between the current pressure inside the seal and the external pressure. When the evacuation mass flow is thus known, the rate of pressure reduction inside the seal can be computed.
-------------------
Henrik Sönnerlind
COMSOL
I studied a similar problem about 25 years ago. It is a very interesting problem, because the apparent seal stiffness (and damping) becomes strongly dependent on the compression speed. By selecting an appropriate size and spacing of the evacuation holes, you can tune the dynamic properties of the seal.
An approach that worked well (experimentally verified) was to not model the evacuation hole, but just solve an extra equation.
The assumtion is then that the evacuation flow velocity is controlled by the difference between the current pressure inside the seal and the external pressure. When the evacuation mass flow is thus known, the rate of pressure reduction inside the seal can be computed.
Please login with a confirmed email address before reporting spam
Posted:
9 months ago
Feb 5, 2024, 2:52 a.m. EST
Updated:
9 months ago
Feb 5, 2024, 2:52 a.m. EST
Thank you, Henrik. You suggestion is super helpful.
To follow your suggested approach, I think I should use the boundary load node instead of the new Enclosed Cavity node released in the latest version 6.2. The "old" approach allows us to define the load freely.
Is using Global Equations (ODE) node a suitable way to compute the pressure reduction rate?
Thank you, Henrik. You suggestion is super helpful.
To follow your suggested approach, I think I should use the boundary load node instead of the new Enclosed Cavity node released in the latest version 6.2. The "old" approach allows us to define the load freely.
Is using Global Equations (ODE) node a suitable way to compute the pressure reduction rate?
Aaron Dettmann
COMSOL Employee
Please login with a confirmed email address before reporting spam
Posted:
9 months ago
Feb 6, 2024, 2:49 a.m. EST
Updated:
9 months ago
Feb 6, 2024, 3:38 a.m. EST
It should still be possible (and easier) to use the Enclosed Cavity node in your case. Assuming that you have formulated your equation of state, which takes the mass flow rate into account, you can enter the pressure load either using the Fluid subnode with Compressibility set to User defined, or using the Prescribed Pressure subnode. The latter subnode is equivalent to using a Boundary Load with the Pressure option. The big advantage of using Enclosed Cavity is that you get direct access to variables for the deformed and undeformed volume (or area), and you can use these variables when defining your equation of state.
The Global Equations node is suitable for solving for any additional global unknowns, for instance an unknown mass flow rate.
It should still be possible (and easier) to use the Enclosed Cavity node in your case. Assuming that you have formulated your equation of state, which takes the mass flow rate into account, you can enter the pressure load either using the **Fluid** subnode with Compressibility set to User defined, or using the **Prescribed Pressure** subnode. The latter subnode is equivalent to using a Boundary Load with the Pressure option. The big advantage of using Enclosed Cavity is that you get direct access to variables for the deformed and undeformed volume (or area), and you can use these variables when defining your equation of state.
The Global Equations node is suitable for solving for any additional global unknowns, for instance an unknown mass flow rate.