INSTAR is a UC Berkeley research group under Professor Dennis Lieu that is designing and manufacturing a high-power, cost-effective flywheel energy storage system for improving electric and hybrid vehicle system efficiency.
I was a member of the lab for 3 years, and directed the lab in 2014 and 2015. During my time in the lab I took part in two large-scale design and manufacturing projects, which are shown below. The platform for this flywheel research is an electric go-kart that the group has also designed and built from scratch over the past six years.
Vibration-Damping Flywheel Mount
The mount had the following design requirements:
- Support the flywheel under extreme driving loads
- Damp vibrations by using rubber isolators
- Safely constrain the flywheel in a catastrophic failure event
The mount was initially developed to support a prototype flywheel, and as such differs slightly from the final product. Some images of the first generation mount are shown below.
Chassis Mounting Pillow Blocks
This first generation mount was revised to account for a new failure mode that I had not anticipated, where the flywheel motor seizes and the entire flywheel and enclosure impart a torque onto the mount. In this mode the rubber dampers would shear, and the flywheel would be unconstrained. To correct this, I created a new design with "seizure blocks" shown below.
Top view of angled impact walls
Rubber Isolators and Containment
Manufacturing was fairly straightforward for this project, as almost all of the parts were designed to be made on an OMAX abrasive waterjet machine. This allowed for accurate hole patterns for the top and bottom plates, and custom contours for the four seizure blocks without having to use a CNC mill. The final product is shown below!
Have more questions? Take a look at the long-form report for this project:
Magnetic Brake Mount
The kart was designed to have an auxillary braking system for stationary testing, where a large mass is attached to each wheel to simulate the kart's inertia and the wheels are spun up and slowed down. Since the friction brakes in the front can't stop the kart when it is on stands, I helped develop a magnetic braking system that attached to the rear wheels.
Another member of the lab did the initial design of the braking system mounts, shown below.
The initial mounts had significant deflection issues, mainly due to the cantilevered load from the chain and the lack of support on the brake itself. I came up with a concept for a redesign that utilized a second plate to form a box-like structure, which would be much stiffer.
I developed this into a working CAD model shown below, and completed detail design on bearing holder and secondary plate. The plate is A36 steel and was waterjet, and the bearing holder is 6061-T6 aluminum and was made on a lathe and a mill. The manufacturing and assembly processes went smoothly, and the magnetic brakes were much stiffer once the redesign was complete.
The redesigned brakes allowed us to capture data on quickly the flywheel could absorb energy from spinning disk that simulated the intertia of the kart.
View the INSTAR Website Here