In our competition the most thermally challenging event is "Endurance".
This event is a 22km drive with a driver change halfway through the event. Keeping the motors
and power electronics cool in the high ambient temperatures of Michigan while keeping the mass of the
cooling system low is a large challenge. The component layout was driven by the fact that the inverters
need a large flow rate of 8-10L/min and the motors can exceed 30PSI. The motors were placed in parallel
to equalize the temperature between them so we wouldn't run into long term performance differences between
the left and right side.
To analyze the thermals of the vehicle and select a radiator I used the Speed/Torque
profile from the teams lap simulation. I overlayed this with the motor and inverter efficiency data
to generate a graph of the amount of heat generated over a lap.
I sourced a radiator which seemed to fit our system. Unfortunately the radiators
we have to choose from had very poor data. Therefore, I decided to do in-house testing on our radiator
at our universities wind tunnel. I did this with a small group for a class project, and found the minimum
windspeed required to keep the components cooled. Which allowed me to design a shroud around the radiator
and size fans to pull air through the system.
Once we had chosen all the components in our system we were able to source a pump that would
match all of the criteria that we had set out. This turned out to be very difficult because our LV rail runs at 24V
which is not the norm for small BLDC pumps. The best pump I was able to source was an underwater pump which meant it
needed to be submerged in a reservoir. The rules of our competition demand that we have a "catch can" to account for thermal
expansion in our system. This actually played into our favour as I was able to argue that our reservoir also acted as
a catch can which meant we no longer had to seperately design a catch can.
The total Pressure drop before motors = 11.6PSI. We will run with the motors at approximately 25 PSI to ensure we do not
damage them through over pressure. Therefore the pump must generate approx 36.6 PSI which translates to 84.44532777367279 ft of head.
Comparing this to the pump I chose we should get approximately 15 LPM of flow which will be more than enough for our system.
Even with more minor losses in our fittings and hoses. The pump we chose is also driven off a physical potentiometer meaning we can tun
the system to find the "sweet spot" and get the exact flow rate we want.
To validate our simulation we had to test our system. This is temperature data from autcross lap tests.
We planned to do endurance testing but were unable to gather the necessary data and decided to validate through an
autocross lap on a cold track day. We compared the thermal generation to the temperature changes to find the thermal
mass of all of the components in our system.
This allowed us to actually graph the temperatures of our components and make sure that at no point
a component spiked above it's maximum temperature to ensure we could succesfully complete our endurance event!
