Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Thermal expansion of a liquid

Matthew Malkin
Hi all,

I’m am trying to produce a model of a fluid enclosed on all sides except one by fixed constraints(walls), on the remaining face a thin diaphragm is in contact with the fluid. As such this is a closed system with no flow.

I would like to model the deflection and stress of the diaphragm wrt the temperature of the enclosed fluid. Clearly as temp increases thermal expansion will deform the diaphragm

I have tried to model the situation using the fluid-structure interface option on 4.2 however I can’t find a tutorial that features a stationary fluid?

Which interface should I be using?

Any help would be greatly appreciated!

Thanks,
Matt

5 Replies Last Post Mar 3, 2013, 9:09 AM EST
Yevgeniy Davletshin
Posted: 5 years ago Feb 19, 2013, 4:59 PM EST
I'm not sure but my guess is that you want to include Heat transfer in your simulations with constant heat source, then in Fluid-Structural interaction you can add from Solid-Mechanics domain -> Linear Elastic Material -> Thermal Expansion.

I'm not sure but my guess is that you want to include Heat transfer in your simulations with constant heat source, then in Fluid-Structural interaction you can add from Solid-Mechanics domain -> Linear Elastic Material -> Thermal Expansion.

Posted: 5 years ago Feb 20, 2013, 1:48 AM EST
Hi

For the solid you can add the thermal expansion, but if your fluid is static in a closed volume you cannot add the "thermal expansion" on the fluid in FSI.

So far I "cheat" usually using only "solid" and define my fluid as a "Linear elastic material" with nu Poisson = 0.49 and E = 1/3 of bulk modulus. Adding in the Nearly incompressible material, to solve also for "p" helps sometimes

But I would appreciate to have a better way, but I'm not sure which domain physics to use under "solid" to correctly model a fluid material of nu=0.5 for thermal expansion (in static approximation, no convective flow, typically for closed volumes filled with a fluid, in a solid enclosure, under thermal load)

suggestions appreciated

--
Good luck
Ivar
Hi For the solid you can add the thermal expansion, but if your fluid is static in a closed volume you cannot add the "thermal expansion" on the fluid in FSI. So far I "cheat" usually using only "solid" and define my fluid as a "Linear elastic material" with nu Poisson = 0.49 and E = 1/3 of bulk modulus. Adding in the Nearly incompressible material, to solve also for "p" helps sometimes But I would appreciate to have a better way, but I'm not sure which domain physics to use under "solid" to correctly model a fluid material of nu=0.5 for thermal expansion (in static approximation, no convective flow, typically for closed volumes filled with a fluid, in a solid enclosure, under thermal load) suggestions appreciated -- Good luck Ivar

Matthew Malkin
Posted: 5 years ago Feb 21, 2013, 6:13 AM EST
Thanks for the replies,

Unfortunately I am still having some problems.

I want to run the model at different temperatures however I am not interested in heat transfer within the model etc

Ivar,

To clarify you are suggesting using the fsi module and then defining both the liquid and solid domain as linear elastic materials? in this case there are no domains in fluid properties.

When I attempt this I get a generic error message regarding geometry and domains.

So far my model (attached) is extremely simple compromising of two domains (solid and liquid) in contact. when I am finalising the geometry should I form a union or a contact pair?

Is there a module I can use to model this other than fsi if I am effectively treating both domains as a solid?

Again sorry for the questions but I am struggling to get my head around this.

Thanks
Matt
Thanks for the replies, Unfortunately I am still having some problems. I want to run the model at different temperatures however I am not interested in heat transfer within the model etc Ivar, To clarify you are suggesting using the fsi module and then defining both the liquid and solid domain as linear elastic materials? in this case there are no domains in fluid properties. When I attempt this I get a generic error message regarding geometry and domains. So far my model (attached) is extremely simple compromising of two domains (solid and liquid) in contact. when I am finalising the geometry should I form a union or a contact pair? Is there a module I can use to model this other than fsi if I am effectively treating both domains as a solid? Again sorry for the questions but I am struggling to get my head around this. Thanks Matt


Matthew Malkin
Posted: 5 years ago Feb 25, 2013, 10:52 AM EST
Any idea's?

Im pretty stuck on this one
Any idea's? Im pretty stuck on this one

Posted: 5 years ago Mar 3, 2013, 9:09 AM EST
Hi

1) in FSI do not use assembly mode, so you get a continuity (in no way a "contact link) between Fluid and Structure
2) in FSI you have no thermal expansion node for the fluid so it will not work like that
3) my proposal was to use all "solid" but this assumes that the fluid is not "flowing" freely, but behaving as a "gelly" incompressible

--
Good luck
Ivar
Hi 1) in FSI do not use assembly mode, so you get a continuity (in no way a "contact link) between Fluid and Structure 2) in FSI you have no thermal expansion node for the fluid so it will not work like that 3) my proposal was to use all "solid" but this assumes that the fluid is not "flowing" freely, but behaving as a "gelly" incompressible -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.