Queries regarding Thermal Analysis of a 4S2P Li-Ion battery pack

Prateek Gaur_20403

Hii, My name is Prateek. I'm a Battery Design and Analysis Engineer. I'm stuck on a small problem and want suggestions from your side.

Problem: There is a 4S2P Li-ion battery pack (18650 cells) that has to be used in the future for the solar-panels motor attachment. There are heating coils of (0.5W) attached on both sides of the cells. so in total, a cell has a heating power input of 1W per cell and 8W for the complete battery pack. The battery pack is inside a box with 6mm insulation (SilicaAerogel Ceramic Fibre Blanket-SACB).

Battery pack Requirements:

Nominal Capacity: 3600mAh

Nominal Energy: 51.84Wh

Output Voltage: 10.8 - 16.4V

Initial Impedance: 110mohm Max (DCR)

Charging mode: CC-CV

Max. Charge Voltage: 16.4V

Charger Max. Charge current: 2A

Charge Temp. Range: 0-60 degC

The target is to find the exact amount of heating power we need to use, to keep the battery above 5 degC during the night in a steady -40 degC environment. The battery will start at 20 degC (the solar panel will operate during the day). Night time in Canada (the location) can be 16 hours in December.

I have tried to perform it in Solidworks but not working as it doesn't even have a cell material library to give battery pack details. Is it possible in the HyperWorks CFD simulator?

Thanks in advance for your time and suggestions and feel free to ask queries.

P.S.: Attached are some photos of the 2S2P pack presently used.

Find more posts tagged with

English

HyperMesh CFD

Accepted answers

Adriano_Koga

Thank you so much for your valuable response, Adriano. You understand the problem totally correct, "The box will be left there overnight and you want to understand the amount of heat to keep it there." Mainly this box with a battery pack is installed behind the solar panels so it will be permanently installed.

The only problem is it's a project for our company and we have to do it. I have SolidWorks but I tried and it's quite impossible I guess by that. So, all I need is that, if Altair Suites is a possible software package for this problem, we will surely go for it. Otherwise, it's a huge investment we will do for nothing (Hope you understand).

Regarding your concern, the company hasn't given us any info about the airflow, so I hope no forced convection. It's a steady cold environment.

So, the only additional support I need from your side if possible is, if you can check that "Optistruct and Activate have the options to give this kind of initial condition of outside temperature, the battery material characteristics (Li-Ion) as well as for insulation. And is it possible to get energy requirements as an output variable?"

I will be highly thankful for your support and time.

Prateek Gaur

in OptiStruct, concerning convection/conduction only, this might help you. It is a very simple model, where the block has convection on the sides.

First model has a fixed temperature at the top surface (20C), and convection with ambient with -40C. For model #1, you can plot SPC Power, which shows how much of power is needed to keep the top surface at 20C through the transient simulation.

Model #2 doesn't have any fixed temperature, and that's why temperatures basically go close to -40C.

Of course the values for material properties were a total guess. But this is only an illustrative model.

transient_block.gif

TRANSIENT_CONV.zip

All comments

Adriano_Koga

this is a really nice study. I understand that this is a passive heat transfer (basic conduction and natural convection) and it has no forced convection, right?

The box will be left there overnight and you want to understand the amount of heat to keep it there.

Honestly, I would try to start from a simplified scenario, maybe using Activate to model the heat exchange in each component, initially in a 1D assumption.

Then maybe going to OptiStruct or HyperWorks CFD to understand the 3D heat exchange.

OptiStruct would be able to give you some directions, except by forced convection. It would be limited to a constant H coefficient for your main thermal exchange. But if this is not so dependent on the "air flow", I believe it would be an interesting starting point.

Now if you have some forced flow, or even if the natural convection itself is important, maybe you will need HWCFD.

There are some tutorials for both of them involving Heat Transfer analysis (transient).

I'd say that you could start with some simple models (maybe the battery only, first) and evaluate the heat transfer. Then, add complexity.

Prateek Gaur_20403

this is a really nice study. I understand that this is a passive heat transfer (basic conduction and natural convection) and it has no forced convection, right?

The box will be left there overnight and you want to understand the amount of heat to keep it there.

Honestly, I would try to start from a simplified scenario, maybe using Activate to model the heat exchange in each component, initially in a 1D assumption.

Then maybe going to OptiStruct or HyperWorks CFD to understand the 3D heat exchange.

OptiStruct would be able to give you some directions, except by forced convection. It would be limited to a constant H coefficient for your main thermal exchange. But if this is not so dependent on the "air flow", I believe it would be an interesting starting point.

Now if you have some forced flow, or even if the natural convection itself is important, maybe you will need HWCFD.

There are some tutorials for both of them involving Heat Transfer analysis (transient).

I'd say that you could start with some simple models (maybe the battery only, first) and evaluate the heat transfer. Then, add complexity.

Thank you so much for your valuable response, Adriano. You understand the problem totally correct, "The box will be left there overnight and you want to understand the amount of heat to keep it there." Mainly this box with a battery pack is installed behind the solar panels so it will be permanently installed.

The only problem is it's a project for our company and we have to do it. I have SolidWorks but I tried and it's quite impossible I guess by that. So, all I need is that, if Altair Suites is a possible software package for this problem, we will surely go for it. Otherwise, it's a huge investment we will do for nothing (Hope you understand).

Regarding your concern, the company hasn't given us any info about the airflow, so I hope no forced convection. It's a steady cold environment.

So, the only additional support I need from your side if possible is, if you can check that "Optistruct and Activate have the options to give this kind of initial condition of outside temperature, the battery material characteristics (Li-Ion) as well as for insulation. And is it possible to get energy requirements as an output variable?"

I will be highly thankful for your support and time.

Prateek Gaur

Adriano_Koga

Thank you so much for your valuable response, Adriano. You understand the problem totally correct, "The box will be left there overnight and you want to understand the amount of heat to keep it there." Mainly this box with a battery pack is installed behind the solar panels so it will be permanently installed.

The only problem is it's a project for our company and we have to do it. I have SolidWorks but I tried and it's quite impossible I guess by that. So, all I need is that, if Altair Suites is a possible software package for this problem, we will surely go for it. Otherwise, it's a huge investment we will do for nothing (Hope you understand).

Regarding your concern, the company hasn't given us any info about the airflow, so I hope no forced convection. It's a steady cold environment.

So, the only additional support I need from your side if possible is, if you can check that "Optistruct and Activate have the options to give this kind of initial condition of outside temperature, the battery material characteristics (Li-Ion) as well as for insulation. And is it possible to get energy requirements as an output variable?"

I will be highly thankful for your support and time.

Prateek Gaur

in OptiStruct, concerning convection/conduction only, this might help you. It is a very simple model, where the block has convection on the sides.

Model #2 doesn't have any fixed temperature, and that's why temperatures basically go close to -40C.

Of course the values for material properties were a total guess. But this is only an illustrative model.

transient_block.gif

TRANSIENT_CONV.zip

Prateek Gaur_20403

Thank you so much for your valuable time and support. I'm highly obliged to you for your help.