# Heat transfer analysis of heating mat

• 916 Views
• Last Post 29 November 2017
• Topic Is Solved
cfd_learner posted this 28 November 2017

HI,

I would like to simulate a heating mat made of heating wire and fiber (cloth). How can I see the temperature rise in cotton/cloth sheet? I assumed cloth as rigid, any porosity etc is neglected. Only interested in heat transfer from heated carbon wire (50 degC) to the cloth sheet (assumed rigid/solid).

How to model this problem and appropriate boundary conditions? A picture is uploaded.

Regards

Attached Files

peteroznewman posted this 28 November 2017

Here is a Steady State Thermal analysis.

A single coil of the thermal mat is created in SpaceClaim in a multibody part.

The hexagonal coil is 1 mm thick and is embedded in a solid that is 4 mm thick.

A mesh that has less than 32K nodes+elements is created to stay within the Student license limits.

The coil has a Temperature BC of 30 degrees C

The front faces of the mat and coil have a convective BC. I just made up a Film Coefficient, you may want experimental data to support this.

The bottom and side faces of the unit square of the mat have a perfectly insulated BC. The temperature is plotted below.

You can pattern the coil to create a model closer to your example with multiple coil segments.
Having multiple coils or coil segments doesn't change anything except the mesh size.

Best regards,

Peter

• Liked by
peteroznewman posted this 29 November 2017

The file extension that was .wbpz had to be changed to .txt because the website doesn't allow a .wbpz file extension to be attached! After you download the file, rename it from .txt to .wbpz then Restore Archive using ANSYS 18.2

In SpaceClaim, there are two bodies in one component and Share Topology was set for the component.

Attached Files

• Liked by
peteroznewman posted this 29 November 2017

If you are new to heat transfer, you could learn the basics of conduction and convection in Module 1 of this free course.

I used "Perfect Insulation" on the four sides to create a symmetry boundary condition to represent what any place that was not near the edge of the mat would look like. If I take my result and copy and paste it 3 times I can get a 2x2 image. Notice that the temperature contours intersect the boundary of the unit square at right angles. That is because no heat is flowing through the plane of symmetry. If I want to see how that is different at the edge of the mat, I could take one side and change that to convection.

As for the bottom face, I set it to perfectly insulated to represent it being in contact with insulation.
I could have set it to convection to represent air being in contact with the back as well as the front.
It just depends on what your model is trying to represent. You can play around with the Film Coefficient to see the effect that has on the result.

In my model, there is almost no temperature gradient through the thickness. If you can make the assumption that the temperature is uniform through the thickness, then you can change from a solid model to a surface model and use shell elements. This will enable you to model a much larger portion of the mat without exceeding the node+element limits.

Cheers, Peter

• Liked by
pgl posted this 28 November 2017

Here's a video showing how to model the heating element/wire:

Here's an electro-thermal actuator example.:

your case will not expand anywhere near as much but the principles used here are the same.

All the best.

cfd_learner posted this 29 November 2017

Thank you Peter. Kindly share the WB project. What actually embed means in geometry modelling? Merge or share topology ?

cfd_learner posted this 29 November 2017

Thank you, Peter, for your prompt response. One more question, why bottom and sides are "Perfect Insulation"? what if all is convective? Will convection coefficient play any role, increasing or decreasing? Actually, I am new to "heat transfer".