Thermal stress and strain

Javaid26 · August 4, 2015, 07:45

Hi All,

I am a newbie at ANSYS and getting confused by a simulation that I ran. I would really appreciate any input on the matter.
I am trying to solve for thermal stress and strain in a lap joint. The joint is placed between two stainless steel plates, the top plate is at a temperature of 470 degree Celsius and the bottom plate is at room temperature. I applied displacement condition on the top and bottom faces of the joint (two aluminium foils and a layer of solder between the foils) with Z-component at 0 and X,Y as FREE.
Pictures are attached. My question is when I run the simulation, the lap joint grows larger in size and ends up with rough edges all over the length and ends. Why is that happening? Is the aluminium expanding and going to that shape because in reality that does not happen. Any thoughts as to what I might be doing wrong??

fresty · August 4, 2015, 08:20

Could you please post a bit more (clear) details, at least the initial/ boundary conditions..
About the lap joint growing larger in size.. not sure about the benchmarks/ validation you have for the analysis, but from your images the total deformation is not larger than 0.005mm, which is reasonably minor... the automatic scale of results always magnifies the resulting plots/ geometric to provide a proportionate view but you could always set the scale to real-time..

Javaid26 · August 4, 2015, 08:45

Quote:

Originally Posted by fresty

Could you please post a bit more (clear) details, at least the initial/ boundary conditions..

Two aluminium foils with a layer of solder are set up as a lap joint. The joint is sandwiched between two stainless steel plates. The top plate is at a temperature of 470 degree Celsius so I set the thermal condition on the top and bottom of the joint as 470 degree Celsius (excluding the area that covers the lap). I have put displacement on the top and bottom faces of the joint with Z as 0 and X,Y as FREE. In Analysis setting tab, I have set number of steps to 5, current step number as 1 and step end time as 1. Then carried out a static structural analysis. That is all.

I haven't validated the results (I will either three dimensional equations off the internet or 1D for that purpose) because first I want to know whether the results should look like this or not. The total deformation is as you said very minor but I am confused about the increase from the initial to the final stage (picture attached at the beginning of simulation).

What do you mean when you say I can set the scale in real time??

fresty · August 4, 2015, 09:06

I understood that bit.. intended to have a better look at the orientation of your model only if there was a good resolution screenshot to distinguish the components (not the camera taken image).. anyways...

About the increase in deformation, as mentioned, the visual is magnified to give you an overall behavioral idea.. you could set the scale factor from the results tool bar at the top, to True Scale (1.0) rather than the scale factor it already has.. This would show you the realistic deformation however the deformation behavior to induced stress would still be consistent obviously..

Javaid26 · August 4, 2015, 09:36

Got your point. Please find a better quality picture after setting to true scale 1.0.

If you don't mind me asking about validation. The equation for stress is alpha*temperature difference*modulus of elasticity. Multiplying these three things give stress value. In the formula, there is temperature difference but I have not put anything during the simulation that would point towards a temperature difference. Also, when I calculated I got 750 MPa rather than 282 MPa that I got from the simulation. Can you please enlighten me on that.

fresty · August 4, 2015, 11:18

You did put the temperature difference while specifying thermal condition as you mentioned earlier and the lower temperature is by default room temperature, however it would be worth explicitly assigning 25 deg C or so to the other faces/ bodies.. Did you try turning on the large displacement option? 282 MPa seems pretty less considering the system that you intend to simulate.. there is certainly a discrepancy probably related to the DOFs.. i might be wrong on this but did you try turning off the Y displacement (set to '0') and observe the results by 1D expansion only...?

Javaid26 · August 6, 2015, 09:31

How much value would you expect in this scenario if 282 MPa is low?
What do you mean by turning ON the large displacement?
I did put the temperature on the layer of solder to 20 degree to have a better understanding. My case is that the joint is sandwiched between the plates so Z direction is set to 0. X, Y are currently free as aluminium can move in these two axes. I can try and do it in 1D which brings me back to my first question, how much value would be reasonable?

fresty · August 6, 2015, 09:52

I wouldn't be able to answer this question as i am not doing the simulation and simply deducing based on the info you provided here.. i believe you should be able to estimate that ..
Nevertheless, as you mentioned previously that by using a simplistic 1D (linear expansion) formulation (del L/L) the stress comes out to be around 750MPa, you should expect similar value in unidirectional/ lateral thermal expansion (so try locking Y direction).. again, this could be one of the ways you could cross check...

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ANSYS Meshing] Error in Workbench "Unable to attach to geometry file"	Jeremie84	ANSYS Meshing & Geometry	70	October 25, 2017 16:03
shear rate and strain rate	Lilly	FLUENT	1	August 29, 2014 00:46
Plane stress and plane strain cases for stress analysis	Sargam05	OpenFOAM	0	May 3, 2013 18:16
What is wall shear stress and shear strain rate?	warlocklw	Main CFD Forum	2	May 21, 2012 06:16
Viscous stress	fluids	FLUENT	0	August 26, 2010 07:57

August 4, 2015, 08:20		#2
fresty Senior Member Join Date: Aug 2014 Location: UK Posts: 213 Rep Power: 12	Could you please post a bit more (clear) details, at least the initial/ boundary conditions.. About the lap joint growing larger in size.. not sure about the benchmarks/ validation you have for the analysis, but from your images the total deformation is not larger than 0.005mm, which is reasonably minor... the automatic scale of results always magnifies the resulting plots/ geometric to provide a proportionate view but you could always set the scale to real-time..

August 4, 2015, 09:06		#4
fresty Senior Member Join Date: Aug 2014 Location: UK Posts: 213 Rep Power: 12	I understood that bit.. intended to have a better look at the orientation of your model only if there was a good resolution screenshot to distinguish the components (not the camera taken image).. anyways... About the increase in deformation, as mentioned, the visual is magnified to give you an overall behavioral idea.. you could set the scale factor from the results tool bar at the top, to True Scale (1.0) rather than the scale factor it already has.. This would show you the realistic deformation however the deformation behavior to induced stress would still be consistent obviously..

August 4, 2015, 11:18		#6
fresty Senior Member Join Date: Aug 2014 Location: UK Posts: 213 Rep Power: 12	You did put the temperature difference while specifying thermal condition as you mentioned earlier and the lower temperature is by default room temperature, however it would be worth explicitly assigning 25 deg C or so to the other faces/ bodies.. Did you try turning on the large displacement option? 282 MPa seems pretty less considering the system that you intend to simulate.. there is certainly a discrepancy probably related to the DOFs.. i might be wrong on this but did you try turning off the Y displacement (set to '0') and observe the results by 1D expansion only...?

August 6, 2015, 09:31		#7
Javaid26 New Member Essex Join Date: May 2015 Posts: 7 Rep Power: 10	How much value would you expect in this scenario if 282 MPa is low? What do you mean by turning ON the large displacement? I did put the temperature on the layer of solder to 20 degree to have a better understanding. My case is that the joint is sandwiched between the plates so Z direction is set to 0. X, Y are currently free as aluminium can move in these two axes. I can try and do it in 1D which brings me back to my first question, how much value would be reasonable?

August 6, 2015, 09:52		#8
fresty Senior Member Join Date: Aug 2014 Location: UK Posts: 213 Rep Power: 12	I wouldn't be able to answer this question as i am not doing the simulation and simply deducing based on the info you provided here.. i believe you should be able to estimate that .. Nevertheless, as you mentioned previously that by using a simplistic 1D (linear expansion) formulation (del L/L) the stress comes out to be around 750MPa, you should expect similar value in unidirectional/ lateral thermal expansion (so try locking Y direction).. again, this could be one of the ways you could cross check...