Drive-train planning
Post date: Apr 2, 2013 5:22:02 PM
The first goal is simply to get the wheels spinning with the car off the ground, without any "polish" :)
During the kart project we used a brushless DC motor from Motenergy, the ME3001, together with a 48V drive from Kelly Controls. For a car, obviously more power is required. A popular option on e.g. www.diyelectriccar.com is to go with a large brushed DC motor, such as the Netgain Warp 9. AC motors are less frequently used, due to higher costs of drives in this power category. Still AC drives (or switched reluctance maybe) are the future; there are no brushes that wear, less EM noise or sparking, higher efficiency, ...
We want to try combining two brushless DC motors from Motenergy, ME0913. The specification per motor is 96V - 12 kW Continuous, 30 kW peak power, with an efficiency of 92%. It is probably possible to increase peak power even more provided enough cooling is provided. 60kW peak would correspond to about 80hp, but such comparison with an ICE engine is difficult because the power delivery is consistent over a wider range of rpm.
A lot of interesting pictures of the inside and modification of this motor can be found at ripperton-electric-track-bike-41173p20.html
The initial plan was to couple the two brushless motors together in line, attached to a mounting plate that screws onto the gearbox. The flywheel is to be lightened and attached on the motor axle using a flange:
By rotating the rear motor by 7.5 degrees, it should theoretically be possible to reduce the torque ripple from about 13% to 4%. Two separate Hall sensors are then needed. You can see this arrangement in the picture above.
The gearbox has mounting holes that do not seem exactly centered. After a lot of searching around we found an excellent description of the bolt and input-shaft locations on the AEVA forums. I made a proper drawing of these measurements in the hope others will find them useful as well:
I also scanned the gasket between engine and gearbox (in multiple parts and the stitched together), but it is not very accurate due to scanner construction:
The main issue with this kind of "in-line dual motor" setup is that the "complete" axle is overconstrained. Thanks to "weber" for pointing that out :) So between the motors a flexible coupling is needed. This makes the whole assembly more complicated, and suitable couplings do not come cheap either. Expecially because of the "non-metric" axle diameter of 22.23mm...