Building on the Tesla fast charge thermal model that was discussed in the Siemens post: Race to the clouds: Battery thermal behavior simulation of a Tesla Model 3. This article looks at the cooling system and impact of a change in cooling direction. Starting with the model and the original Fast Charge Scenario.
The original analysis set out to Simulate the Pikes Peak Hill Climb with a Tesla Model 3 and check in detail the battery thermal behavior.
- Electric model: regarding the electric part, we are considering a Dynamic Equivalent Circuit Model (ECM) using a pre-calibrated model from the Simcenter Amesim database as illustrated in Fig. 3. We selected the NCA High Energy cell model, with its cell capacity resized to 4.8 Ah
- Cooling model: regarding the cooling part, elements from the Thermal and Thermal Hydraulic libraries have been used and associated to the electric component through the thermal port.
Fast Charge Scenario
The first scenario of this battery thermal behavior simulation with a Tesla Model 3, is considering a fast charge scenario:
- 2C fast-charge, i.e. 150kW of charging power to charge the battery pack from 20 to 75% state of charge (SOC) followed by a decreasing current (~constant voltage) till 100% of SOC
During this fast charge, we fix the coolant flow to 30L/min at an inlet temperature of 25°C.
The results presented on the above figure show a maximum temperature of about 36.3°C for the modules 1 & 4 and 39.3°C for the modules 2 & 3, which have 2 additional bricks in series. We reach this maximum level at the end of the 2C charging period.
The maximum outlet coolant temperature is close to 37°C
According to these results, we can see that the maximum cell temperature is below the 45°C limit. The cell-to-cell temperature difference is however reaching 12°C, which seems to be at the upper limit.
Change in Cooling Direction
The cooling in the Tesla Model 3 runs front to back. Each module connected in parallel with the coolant inlet.
This is as per the Tesla Model 3 battery design and as per the original analsysis. However, we wanted to see what the impact would be of running the coolant on the shorter path across the pack.
This is a simple schematic and quite easy to build within the model and simulate. However, in reality this would have a significant impact on the overall design and package. The simulation shows the temperature delta (cell to cell maximum temperature difference) for the pack in fast charging reduces by ~2°C.
An alternative is to flow the coolant from the mid-point and then out to front and rear. Although this would require some significant changes to the design, it is easier to see how this could be achieved.
Above is a simple schematic of the flow path relative to the modules. Below is an image of the Simcenter Amesim model:
The result is a much lower cell temperature difference across the pack.
The following image shows the temperatures in each block of cells for module 1 for the 3 designs.
All of the models were developed in Simcenter Amesim. This allowed the existing model to be modified for different cooling options very quickly and re-run to compare certain parameters. In this case the focus has been on the temperature difference of the cells across the battery pack. The optimisation could have easily been to minimise the charge time.
When designing a battery pack the ability to look at different thermal system designs is very powerful and can give you data to enable your decision making process.