Thermal stresses in microchannels - 3-2-1 method

[Cathie16]

Hello all,

I am relatively new and would like to clarify some doubt in analyzing thermal stresses in micro-channels of an heat exchanger.

I just consider only a small cross section of the entire model to simplify the model since my interest is only inside the channels.

In this case where would be ideal to constraint (boundary conditions)?

I usually constraint in the lower corner and go further away from this point to analyse the results. But is there any better way ?

Thanks for your time and support in advance,

Regards

 

[karachun]

If your channel can expand freely (as usually heat exchanges do) than you can use 3-2-1 method described here. If you apply constraints properly you get zero or near zero constraint forces at constrained nodes.

https://www.digitalengineering247.com/article/free-floating-fea-models

Recently there was a big discussion about thus method in other thread.

https://www.eng-tips.com/viewthread.cfm?qid=516074

UPD. Look like ti is not so easy. If you cut small part of crossection than those cut planes should remain planar afrer analysis, or not? Is is possible to calculate whole crossection of heat exchanger? Like mesh this small part and copy paste mesh and connect coincident nodes.

 

[Cathie16]

Hello Karachun,

Thanks for your reply. Its indeed a great learning for me to learn about this method since I principally work with pressure vessels and always fix the bottom which is not reasonable. I will analyse this in depth.
For this model, I am only concerned about the thermal effects on the channels due to the high gradient in temperature between them. And hence I believe considering only the small part of cross section should be a valid consideration.
Thanks again,
Have a great day

 

[Cathie16]

Hello Karachun,

I have tested 3-2-1 method and the results are quite satisfying. The ideal loading case for the problem I have explained above is applied pressure in all the cavities of both circuits along with the temperature gradient between them. I wanted to check if only with temperature loading if there will be only deformation which should be the case for the model that is free to expand in all directions. But is it possible there can be some constraints created due to the restriction in expansion due to the temperature gradient between them?

Thanks in advance,

Regards

 

[karachun]

Run your case and check reactions at constrained nodes. If they are zero that nodes do not prevent model expansion because other vice you get some reaction forces.

 

[Cathie16]

Thanks again. I found the problem. For uniform thermal loading I get VM almost zero and some translation based on the thermal expansion co-eff. This method is the best for my application and thanks again to suggest this !

 

[karachun]

Yep, parts only expands/shrink due to uniform temperature load without stresses if nothing prevent expansion. You should introduce some nonuniform thermal load to get thermal stress in model constrained by 3-2-1 method.