Tachyon Battery

Overview

2018

CalSol is Berkeley's solar car team. The team builds a new car every 3-4 year but races the most up-to-date car annually. Tachyon was the team's first attempt at a multi-occupant vehicle, with all previous cars being single seaters. This meant a lot of things would need to be bigger, including the battery.

I took on the task of designing the battery after the car had gone through some initial design decisions from previous members and an initial battery design that got scrapped. These decisions were:

• Pack Configuration: The pack needed to be 36s36p. This provided good battery "size" and working voltage for the motor controllers.

• Pack location: The pack needed to behind the rear seats, on top of the car's tunnel. The car is built like a catamaran. Ideally the pack would be low and centered, but this got overlooked by the composites team.

• Cells: LG MJ1 18650. These cells provided the best energy density of 18650s on the market.

Design (Old)

Design flaws: 

• The modules themselves are designed from fiberglass sheet which made prototyping difficult. All iterations had to be outsourced.

• The busbars were made of copper which is heavy. Not shown are the copper interconnects which added even more weight. 

• A lot of empty space between modules meant air was more likely to flow between modules than through cells. This could cause overheating issues.

Design (New)

Design changes:

• The redesigned modules used polycarbonate which could be waterjet in the student shop. It is also relatively cheap.

• The redesign busbars used 6063-O aluminum which ended up being 9 pounds lighter in  after adding up all busbars and interconnects

• The redesigned pack structure minimizes empty space between modules for better cooling. The pack never overheated at its first race in Austin, TX where ambient temperature peaked at 96 degrees F!

The Race

I tried to keep the overview as brief as possible, but the design and assembly was a long 6 month process that started with learning how to use Solidworks and ended with a successful race at FSGP. The car completed over 250 miles worth of laps in the course of three days and only overheated once due to a faulty thermistor reading.