SimScale CAE Forum

Heat Simulation in Vacuum


#1

Hi there,

If I want to simulate the temperature and the stress in a object which is in vacuum, one side will be heated via radiation from a source and the other sides are at low temperature, do I have to consider special settings?


#2

Hi @sburger!

You can do a thermostructural analysis in this case. On one side apply a surface heat flux and rest should remain at the initial low temperature. But of course you can play with other parameters also. You also have to constraint your body as this analysis also take in to effect structural deformations.

Good luck!
Ahmed


#3

Hi @sburger,

That sounds like a very interesting project. I agree with @ahmedhussain18, a thermostructural analysis sounds like the best analysis available if you want both temperature and stress fields. If the body has a high conductivity, the body can heat up a lot, causing it to change in shape, and/or resulting in unresolved stress concentration. In such a case, the deformation of the body is also likely to be non-linear, and hence I would suggest you to take non-linear effects into consideration.

But if the body does not have a very high conductivity, and you don’t expect it to change much in shape, I would suggest you perform a Heat Transfer analysis. Since you have not mentioned any structural load is being applied in the simulation, I do not think you need to focus on the stress field.

Also, it is good to keep in mind that the simulation is an Uncoupled thermomechanical analysis. That means that the temperature field is first solved, and then the structural analysis follows, with the temperature field solution being used as a load for the structural analysis. As many materials change their material properties at higher temperatures, acting generally as more viscoelastic in some cases, an uncoupled analysis may not always lead to excessively accurate results.

Hope this helps.

With regards,
Vikrant Srivastava


#4

Hi @v_srivastava and thank you for your detailed input here!

Just wanted to add something. As mentioned by @v_srivastava, thermostructural analysis can be solved either in the way he mentioned (uncoupled thermomechanical analysis) or in another way. The second possibility would be a coupled thermomechanical analysis:

In that case, the temperature distributions caused by the cooling are calculated simultaneously with the current stresses (in one step). You have to chose this analysis if you have non-linear material behaviour or non-linear boundary conditions in your simulation. In those cases you can be sure that during the cooling, stresses will affect the temperature distributions and vice versa.

An example would be phase transformation of a material due to heat generation that is caused by friction.
In that case, exact time-dependent material properties have to be known.

Best,

Jousef


#5

Hi @jousefm,

Is the possibility to use a coupled thermomechanical analysis available (to non-paying users) on the Simscale platform currently? That would be very useful. I went through the Thermomechanical analysis in the “Analysis Type” and couldn’t find Coupled Thermomechanical analysis option, and hence I didn’t include that in the answer above.

Nevertheless, if the option is available, I agree that coupled thermomechanical analysis would be the best analysis for the given problem.

I hope you have a great day ahead!

Regards,
Vikrant Srivastava


#6

Hi @v_srivastava,

thanks for mentioning this! I just added my answer to have everything covered, for thermomechanical analysis at least.

This feature is not implemented yet but you can add a request here: Feature Request

If there is a keen demand for this feature, the priority of it will increase and chances are higher that this feature will be available on SimScale in the near future.

P.S.: Paying users as well as non-paying users have the same features listed in their platform. Just to be sure users do not get confused here. :wink:

For more details, please see: SimScale - Plans & Pricing

Thanks and enjoy your week!

Jousef


#7

Hi v_srivastava,
Thank you for you detail information.
May I ask at which temperature range for metal I can expect your mentioned change of material properties to more viscoelastic?

Best regards,
Stefan


#8

Hi @sburger (Stefan) ,

The properties depend on the metal chosen. There is no one standard temperature range for the metals, but as a rule of thumb, you can expect it to be viscoelastic near the temperatures when there is a phase change.

Regards,
Vikrant Srivastava


#9

Hi Vikrant,

Ok. To concrete it, because I couldn’t find anything in the internet, which is the temperature for steel, copper and aluminium?

Regards,
Stefan


#10

Hello again Stefan (@sburger),

I am sorry I can’t be more specific, but there is no possible way to tell a single temperature at which the metal would behave like a viscoelastic. I can tell you that you can use the glass transition temperature of each metal, but it would be misleading.

The problem lies in the fact that viscoelasticity is a property defined by the atomic bonds of the element. The amount and purity can substantially vary the viscoelastic behaviour of the material. The age of the material can also effect its viscoelasticity (like it does in the skin). Many materials which are elastic at room temperature behave as viscoelastics when heated (Jet engine blades), and hence the temperature (and also entropy/number of free electrons) can change the extent of viscoelasticity. Aluminium, at room temperature, is very un-viscoelastic.

The exact- or even approximate- values can only be found out after extensive experimentation. Unfortunately, there is no other simple way.

I hope this was helpful.

With regards,
Vikrant Srivastava