Saturday, September 11, 2010

E-transmission refinements

I now have over 100 miles on the 4 turn /2 turn system.
The most important operating parameter is not to switch when there is drive current in the phase wires. I designed a simple system to detect throttle position using a micro switch on the throttle body

When in the zero throttle position the switch is closed
 allowing winding switching

I also added momentary push button switching with the use of a latching control relay

The next feature I want to add is speed sensing such that one cannot down shift to a 4 turn winding if the speed is to great.

As I said I have over 100 miles on the system and the data supports the early claims, improved battery matching is the main benefit, so why do it.

  1. Lowers the needed C rate from the batteries, less demand on the batteries this means for the same performance off the line the batteries need to deliver half the amps. "Why" motors generate forward force by amps times the number of turns hence the term amp/turns, the increased number of turns works like having more amps. The down side is higher winding resistance, which is why we switch back to a two turn winding for high speed operation.
  2. Better low speed torque, because controllers can deliver a limited amount of amps, so as long as we aren't voltage limited (low speed) then it works as if the controller has twice the amps. In reality we are using twice the amp/turns
  3. It is possible to down shift while in Regen mode and it is a cool feeling just like downshifting a transmission. What happens is this; if one is in Regen, in the two turn mode and you pop it into four turn the Regen doubles very cool feeling, it's just like downshifting and letting out the clutch and getting engine braking.
So is all this complexity worth it, Stay tuned for the answer to that question.


  1. Why not let the speed switch kill the throttle signal for a brief moment while doing the switch, or does that turn the controller into regen ?

  2. That can be done but some small delay would be needed when you switch the motor winding count. The motor would need a few hundred milliseconds of being off before the winding relays switch, all doable. This would lead to automatic switching with speed and load being the determinants. Adding the control logic - microprocessor control is a independent project from proving out the concept of a two speed in wheel motor.