Cylindrical Cell Thermal Model

[ChaddKiggins]

I am trying to create a thermal model showing the spatial variation of the temperature during high rate charge events for a cylindrical cell. I mainly want to capture the temperature variation across the radius of the jelly roll. I expect the center of the jelly roll to be hotter than the exterior, but want to be able to quantitatively show this through simulation.

When using the "x-full" thermal model, it appears it is solving the x-spatial dimension of a "unit cell" (anode - separator - cathode, i.e essentially what I discretize for the DFN model). Am I able to extend this to the entire radius of the jelly roll with the currently available thermal models?

I did come across this very nice paper where PyBaMM was coupled with OpenPNM to create a 2-D thermal model of a jelly roll. Would this be the best method to simulate the spatial temperature distribution of a jelly roll during high rate charge events (or any other charge/discharge events for that matter)? I do have access to the necessary equipment to carry this out, but wanted to ask for opinions.

I plan to carry out the work on a few SDI power & energy cells and am certainly open to any suggestions or recommendations.

[TomTranter]

Hi @ChaddKiggins thanks for the question and comments on the paper. The code for that paper is in a separate repository and it's been on my to do list to make a public release. We're working more generally on distributed pybamm models such as packs too and the plan is to manage the solving directly in pybamm in the future but other projects have taken precedence. My colleague @rtimms also made a radial thermal model which doesn't simulate the whole cell but does allow you to resolve temperature distributions in the radial direction. Correct me if I'm wrong Rob but are all the sections assumed to be held at the same voltage? Reasons you might want to consider a 2D or 3D models rather than a 1D temperature distribution is considering variation of current distribution and degradation at specific locations like tabs.

[rtimms]

There are different limits you can take depending on which regime you are in. The one most relevant for battery materials is where the potentials in the current collector vary as a function of radial and axial position only (to leading order). In this limit, the potential drop as you travel some small angle around the spiral is small compared with the drop when you have completed a full turn. You can in principle couple this model to a thermal model that is 3D with an anisotropic conductivity, but things can break down if you get large temperature variations in the azimuthal direction (e.g. due to tab location).

If you're happy to get out a 2D potential and temperature profile that only depends on radial and axial position it works really nicely. There is a simple example notebook here and the paper is here.

[ChaddKiggins]

Thank you for your detailed responses and the excellent resources, @TomTranter and @rtimms. The info helps out a lot and provides a solid starting point for my project. The cells I'm working with typically have a single anode/cathode tab, which should align well with the models you've presented. While the cathode tab is sometimes located within the jelly-roll, I should be able to account for this.

Ultimately, I will be exploring tabless designs. It's great to see that much of the modeling approach will be applicable to this as well.

Being relatively new to modeling and simulation, I really appreciate all the documentation/notebooks and the work you guys do.

[lalith-sumanth]

@ChaddKiggins I'm also trying to create a full thermal model for cylindrical cell. Can you please share your work, it will be very helpful.

[TomTranter]

The repo is here https://github.com/TomTranter/JellyBaMM but it requires pinned versions of some of the dependencies so make a new environment to test it

I will be updating it in the future

[ChaddKiggins]

@lalith-sumanth, feel free to reach out and I'd be happy to share and chat about it. You can find my email in my bio. I have been using a few tools to resolve the 3D temperature distribution in a cylindrical cell. They are fairly simple and make a few assumptions but get the job done. Soon, I definitely want to begin playing around with the JellyBaMM repo as it has many more fantastic capabilities.