Regenerative Braking

Regenerative braking discussion

Rich Sadler, 2008.01.22

How will the bike wheel drive the electric motor when you're
regenerating? Like I said, you can't have a derailer in the drive train
to the motor if the wheel is to drive it- the chain will knot up. Won't
you have the same voltage limitation as with the battery- when the
capacitor is charged to the same voltage the motor-as-generator is
outputting, it won't take any more. Or do you have transformer or
DC-to-DC converter to up the voltage? Maybe if you have multiple
capacitors or batteries, you could switch from one to another as each
"filled up". If John & I were to convince you of the error of your way &
you went with a freewheel system, you can pedal backwards to drive the
motor with human power when coasting & at stop signs to generate

I also was confused by your post on cranks- don't you need tandem cranks
with the ability to mount a chainring on the left crank? And the
expensive system also has freewheels built into the crank axles so that
one rider can stop pedaling if s/he chooses while the other pedals on- I
thought that requirement was in your spec.

Thanks for the clarificiation. We are considering replacing the battery
with a capacitor and running the regeneration from braking and some
pre-ride charging. Thoughts?

Eric, 2007.12.30

Clutch sizing will be based on torque transferred. If we wish to
decouple the passenger, the clutch should be placed before the
transmission, and like wise for the motor generator. Did people lose
interest in leaving the motor connected all the times and allowing
freewheeling by disconnecting the field windings?

Would we like electromagnetic? It could tie in nicely depending on if/
how motor controller (or vehicle logic) is constructed.

Dr. Layton - how much can the average human apply to the pedals (as it
is a recumbent - standing weight is not applicable) and how long are
bicycle cranks (a range is fine)? I do not have the bicycle science
book in front of me to look it up, I thought you may have an ideal
from experience.

Brad, 2007.12.29

I would like to pursue both the throttle and torque sensor options in
parallel. Please do continue with your throttle design. Vinay Gadia is
pursuing the torque-sensor design. The motor selected should be the same
in either case. The torque-sensor design exists on the eBike and
probably on the Go-One as well. I agree, it is more difficult to
implement, and does present control issues, but allows the driver's
hands to be used only for steering and braking.

We will need a clutch to disengage the motor from the drive axel.
Perhaps this same clutch could disengage the rear rider as well.

Joe P, 2007.12.22

To clarify a little on Ray's original message with respect to
regenerative breaking and the "neutral gear".

Regenerative braking will only occur when the brake is engaged and a
load (some impedance) is seen at the terminals of the motor. This will
only occur when the user physically (or electronically) engages the
regenerative brake. The unwanted component that Ray refers to as
"regenerative breaking" is actually the internal mechanical resistance
associated with the motor spinning freely (i.e. no voltage or load is
applied to the motor).

If you would like a demonstration of this affect do the following with
the bike we used for the Innovate or Die Contest in Hess:

Step 1: Remove the belt to the motor and begin peddling. Note here that
bike peddles "easily" with little to no resistance.

Step 2: Now replace the belt to the motor and remove the motor's
electrical terminals (alligator clips) from the recharging circuit
board. Again, begin peddling. Note here that the motor adds some small
resistance to the peddling so the bike is slightly more difficult to
peddle. This is what we would like to remove by adding the neutral gear.

Step 3: Connect a 100 watt light bulb to the motor's terminals. Begin
peddling until the bulb glows. Here note that the bike provides a
tremendous amount of resistance to peddling as compared with what was
seen in steps 1 and 2. This is 'true' regenerative breaking. The
resistance is the system requiring extra energy to power the light
bulb. (In the case of the car the energy would charge the capacitors).

To address your question: "Why would you ever want to run in neutral?".
You would want to run in neutral anytime you do not want to slow down,
nor do you want to use the electrical motor. I would anticipate the
follow 2 scenarios. First you are peddling and maintaining velocity and
second you are costing and do not wish to accelerate or de-accelerate.

To address your second question: "Would it be when the vehicle was
delivering too much power to the super capacitors?"
No, for several reasons. First if the regenerative breaking system is
designed properly then there will be ample capacity in the capacitor
bank. Secondly if too much power were to be transfered to the capacitor
then the protection circuit would handle it. There are many protection
circuits that can be here the two most common are 1, dissipate the
excess energy in heat or 2, open the circuit. The advantage to
dissipating the energy as heat is that the motor will still work as a
regenerative brake but this also adds mass as heat dissipation elements
need to be included. The advantage to the open circuit is that it can
be implemented without adding much mass but the motor will cease working
as a break as the motor will operate without an electrical load (refer
to the motor spinning freely above. To determine which mode of
protection is best we first need to determine the amount of energy that
the regenerative breaks can capture and compare that the mass penalty
associated with heat dissipation elements.

Brad, 2007.12.22

I understand the issues of the regenerative braking especially if you do
not want it on all of the time. What about a clutch that is designed to
be by default engaged during driving and regen, but opens when you want
to run in neutral. Why would you ever want to run in neutral? Would it
be when the vehicle was delivering too much power to the

I think the separate throttle is fine, but Vinay is working on an input
to the motor that basically amplifies the output from a torque sensor to
signal the motor when high power is required. Vinay, how is this coming?
If this ends up being too complicated, the throttle is an okay fall

Ray, 2007.12.22

Regarding the regenerative braking, transmission, and drive train, we
came up with the following ideas and are looking for feedback:

- For regenerative braking to work properly, the motor must spin with
the axle. However, we do not want regenerative braking to occur at all
times, so there must also be times when the motor does not spin with the
axle. This will be accomplished through the use of a "neutral" gear.
Is there a better way of doing this?

- For the transmission on the electric motor, a bicycle-type gearing
will be used, controlled by a servo motor acting upon feedback of motor
and wheel speed, a variable resistor to determine what gear the vehicle
is currently in, and a throttle input.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License