Beginner's troubles

General discussion about Elmer

Beginner's troubles

Postby Tonio » 13 Mar 2017, 23:56

Hello,

I'm new to Elmer software that I downloaded two days ago.

I wanted to start with a very simple case: Compute the transient temperature profile through an homogeneous material in response to a power transient applied through the upper surface.

Conditions are straightforward: Constant temperature on the bottom surface, perfectly insulated side surfaces, power transient applied through upper surface. Body to simulate is 1mmx1mmx0.28mm.

First issue: I used FreeCAD to design this simple shape and imported into Elmer. No issue to import and mesh it, but units have been lost in the process, so my shape was imported as a 1mx1mx0.28m which obviously increases by 1.0e6 the transient thermal response time. Any way to restore the original geometry? FreeCAD will not let me input a dimension of 0.001m, will replace it by 1mm.

Second issue: I struggled to insert the input heat flux condition through the upper surface. I thought first that units should be W/m² as it is defined as an heat flux through a 2D boundary. I realized digging into the model documentation that Elmer expects a mass heat generation in W/kg. How should I convert my input transient surface flux in W/mm² into the unit expected by Elmer? I can convert to W/m² easily, but dividing by material density in kg/m3 is nonsense as it would lead to W/kg*m.

I hope some of the experienced users on this forum will be able to help.

Antoine
Tonio
 
Posts: 8
Joined: 12 Mar 2017, 20:35

Re: Beginner's troubles

Postby dmitry » 14 Mar 2017, 09:33

Hi!
First issue: You can scale the mesh in ElmerGUI, just paste in
Model-->Setup-->Simulation-->Free Text
following instruction
Code: Select all
Coordinate Scaling = 0.001


Second issue: Heat flux q=k*dT/dn is in W/m² (Eqn. 1.18 in Models Manual). Anyway, you can use any unit system. But you have to rescale all the material properties by the hands - Elmer has no unit control system. But I ussually work in SI.
dmitry
 
Posts: 47
Joined: 09 Dec 2014, 16:52
Location: Russia

Re: Beginner's troubles

Postby Tonio » 15 Mar 2017, 01:53

Thank you so much for your help.

Coordinate Scaling fixed my first issue.

I have now correct dimensions for my test setup, and heat flux of 1W/mm² through upper surface gives me a steady state temperature consistent with hand calculations. Going now to transient heating and more accurate geometrical model.

Thanks again

Antoine
Tonio
 
Posts: 8
Joined: 12 Mar 2017, 20:35

Re: Beginner's troubles

Postby dmitry » 15 Mar 2017, 16:47

You're welcome! ;)
dmitry
 
Posts: 47
Joined: 09 Dec 2014, 16:52
Location: Russia


Return to General

Who is online

Users browsing this forum: No registered users and 1 guest