Thermal stress analysis

Do you have 2 or more different materials?

Nope, I used single material. So there should not be any stress but it shows.

Solver does give warning in the .out file.

 *** WARNING # 1284
 No SPC or SUPORT1 data specified for static SUBCASE 1.
 Make sure the model is constrained otherwise.

Theoretically, when you make a Thermal Analysis, you need ONLY Thermal Boundary conditions. In this case, all Mechanical Boundary Conditions are not needed.

@Rahul R

I also tried without AUTOSPC, but got the same results.

@Q.Nguyen-Dai

Is it like inertia relief analysis?

FYI

<?xml version="1.0" encoding="UTF-8"?>

Hello Rahul,

The equation u mentioned for strain is shows total strain due to thermal loading & mechanical loading. However such analysis comes under Coupled Linear Heat Transfer/Structure Analysis example OS-1080.

If we can consider such both loading (Thermal & mechanical) in thermal stress analysis then what is difference between thermal stress analysis & coupled Linear Heat Transfer/Structure Analysis theoretically ? 

As we know there is no mechanical loading so theoretically stress generate by mechanical loading should be zero. As well there is no restrictions to expand or contract the material theoretically stress due to thermal loading also should be zero.

mentioned equation for stress can be divide into two part thermal & structural stress.

<?xml version="1.0" encoding="UTF-8"?>

https://www.youtube.com/watch?v=JCEIQqcmYGY

https://www.youtube.com/watch?v=aoZ3P030loI

Hello Rahul,

The video link provide u also justify that if there is no BC then stress should be zero. 

At last of video he defined two equations one is validated when support (con.) not yield, another one when support yield. However, no explanation for stress without support.

It may error in coding. Plz check further and let us know.

I think software is giving stresses because of temperature difference. In Fem file you can check the temperature defined on grids. If you run same model with same temperature (let say50 ) you would find the difference in thermal stress as negligible. 

Hello Rahul,

Only Temperature difference never cause of stresses, it should be constraint to develop stress according yo basic physics.

If applying higher temperature value still no significant differences in stress then it is more confusing.. stress should be temperature dependent.

Agreed with @GAJENDRA KUMAR NHAICHANIYA, different temperature can make stresses, but not always!

If you have only ONE material and you have no Mechanical constraints, you can not have stresses due to different temperature. You have only deformation due to different temperature, but never stresses!

I just found one document in web with search for thermal stress analysis for unconstrained structure. PFA same and read summary on last page point 4.

4. Thermal stresses may be introduced to the solid if:

(a) the induced solid is physically constrained  with a uniform temperature change in the solid

(b) Unconstrained solid but with temperature  gradients in the solid. Unconstrained solids with uniform temperature change does not generate thermal stress.
 

Unable to find an attachment - read this blog

Hello Rahul,

Totally agreed with your answer this time. But still in my problem i applied constant temperature throughout the model & stress are developed. I still confused why it happened?

I guess the magnitude of stress will be negligible.

Nope, as we increase temperature value stress value increase significantly.

In one case I found

For 50 C,  stress =40.57 mpa

For 500 C, stress= 332.8 mpa

Thermal expansion/strain/stress are calculated based on the thermal coeff. and temperature difference.