Monday, May 15, 2017


Following this year's FSAE season, I had the burning desire to make something "fun" again. No more messing with the Sevcon until dawn, or trying to convince other people to show up and do some work, or getting bogged down by schoolwork (not that it really stopped me from working on car).

Of course, this resulted in taking things way too seriously. As a result of planning out the car team class' curriculum for next fall, all I could think of was the meat of the design process. Designing to spec, meeting a performance goal, taking into account your budget, getting to make something cool to get around a problem.

And so, I met myself somewhere in the middle - an electric go-kart. In partial fulfillment of my desire to whip around the new engineering building at Tufts in style, and to have a working electric vehicle again, I started on the aptly named Tee-Mobile about three days ago.

Partially Transparent-Mobile

I started with a performance goal and a budget- hitting 20 mph in around 10 seconds, but also spending less than $600. So, I whipped out a spreadsheet and used some good ol' idealized physical models to get me some enumerations.


From my mechE friends, I've been told that when you start designing a vehicle, you start with the tires, then go to the suspension and chassis, and then the powerplant you need to hit the numbers you want.

I of course, was granted the luxury that there was almost no chance that I'd be doing burnouts with the current state of cheap scooter-worthy parts, so contact points weren't an issue. (I'd need something upwards of 22kW to get a burnout going from standstill)

I had no intention of having a suspension given the terrain would be entirely smooth, and I didn't have that many restrictions on the chassis - a fairly small turn radius would be nice, however the primary concern was whether I could fit on it or not.

So, I got to skip a lot of the hard stuff and went to pick out components.

Quickly, I discovered that this would be kind of hard to pull off.

The game was essentially "find the low end torque" - not especially hard with electric motors, but kind of hard with hobby airplane parts.

Starting with the motor:

I settled on the Turnigy SK3 6374.

Has a kV of 192 RPM/V, and is rated for 2.75 HkW (Hobbyking kilowatts). Alternatives included some offerings from Alien power systems, and other wholesale RC sites, but the 6374 wasn't so fast that I couldn't gear it down sufficiently, nor was so large that it couldn't fit in my preferred u-channel vehicle construction method.

I also wouldn't feel too bad putting ~4kW through it for short stints. Cooling will definitely be negotiated, with an auxiliary 12V supply on board to power fans, but also fancy lights and the like.

Once that was settled, I could then move forward with futzing with the steering geometry. Run of the mill Ackerman was chosen because we're going to be spending most of our time at 'low speeds' and I'd like the wheels to last more than 5 turns.

Going in circles: no longer a bad thing

Other niceties include an actual chain tensioner that has a spring. Having learned my lesson from the scooter, I have also opted to include all four mounting posts for the motor.

You may have also noticed the lack of brakes: having gotten sick of doing mechE things, and not enough of the zappy things (although in the spirit of the sloth, I may even avoid that), I opted to go for the latest and greatest in cheap Chinese motor controller technology, which implements not only regenerative braking, but also SINE DRIVE. Let's just hope I'm not rolling down a hill when the batteries are already dead. 

Big controller or small cart?
For now, the 1.5kW, 18 FET version of the latest Chinese brushless motor controller was chosen, but we may downsize and hope that whatever FETs in the smaller, 12 FET variety are equally capable of being overdriven via forcible resizing of the current shunt.

To be continued!

No comments:

Post a Comment